Regional association plotting of 11 loci associated with CAC.
A ‘druggable-MI-targets’ project
Sander W. van der Laan, PhD | @swvanderlaan | s.w.vanderlaan@gmail.com
2022-06-30
Setup
We will clean the environment, setup the locations, define colors, and create a datestamp.
Clean the environment.
Set locations and working directories…
Create a new analysis directories.
- general directory
- for plots
- for output of summary results
- for baseline tables
- for genetic analyses
- for Cox regression results… a package-installation function …
source("scripts/functions.R")… and load those packages.
source("scripts/packages05.R")If prompted for GitHub credentials, enter your PAT in the password field
Using GITHUB_PAT from Sander W. van der Laan (credential helper: osxkeychain)
Using github PAT from envvar GITHUB_PAT
Skipping install of 'patchwork' from a github remote, the SHA1 (79223d30) has not changed since last install.
Use `force = TRUE` to force installation
Using github PAT from envvar GITHUB_PAT
Skipping install of 'RACER' from a github remote, the SHA1 (1394c9d4) has not changed since last install.
Use `force = TRUE` to force installationWe will create a datestamp and define the Utrecht Science Park Colour Scheme.
Today = format(as.Date(as.POSIXlt(Sys.time())), "%Y%m%d")
Today.Report = format(as.Date(as.POSIXlt(Sys.time())), "%A, %B %d, %Y")
source("scripts/colors.R")Introduction
We will parse the data to create regional association plots for each of the 11 loci.
Setting the NPG colors
library("scales")
pal_npg("nrc")(10) [1] "#E64B35FF" "#4DBBD5FF" "#00A087FF" "#3C5488FF" "#F39B7FFF" "#8491B4FF" "#91D1C2FF" "#DC0000FF" "#7E6148FF"
[10] "#B09C85FF"show_col(pal_npg("nrc")(10))
# show_col(pal_npg("nrc", alpha = 0.6)(10))Load data
European and African-American
We need to load the meta-analysis summary statistics from the European - African-American ancestry analysis first.
gwas_sumstats_racer_EA_AA <- readRDS(file = paste0(OUT_loc, "/gwas_sumstats_complete_racer.EA_AA.rds"))European-only
We need to load the meta-analysis summary statistics from the European-only ancestry analysis.
# gwas_sumstats_EA <- fread(paste0(GWAS_loc,"/CAC1000G_EA_FINAL_FUMA.txt.gz"),
# showProgress = TRUE)
# names(gwas_sumstats_EA)[names(gwas_sumstats_EA) == "Pos"] <- "Position"
# saveRDS(gwas_sumstats_EA, file = paste0(OUT_loc, "/gwas_sumstats.EA.rds"))
#
# gwas_sumstats_racer_EA <- subset(gwas_sumstats_EA,
# select = c("MarkerName", "rsID", "Chr", "Position", "Pvalue"))
#
# saveRDS(gwas_sumstats_racer_EA, file = paste0(OUT_loc, "/gwas_sumstats_racer.EA.rds"))
# rm(gwas_sumstats_EA)
gwas_sumstats_racer_EA <- readRDS(file = paste0(OUT_loc, "/gwas_sumstats_racer.EA.rds"))African-American-only
We need to load the meta-analysis summary statistics from the African-American-only ancestry analysis.
# gwas_sumstats_unfiltered_AA <- fread(paste0(GWAS_loc,"/CAC1000G_AA_FINAL_FUMA.unfiltered.txt.gz"),
# showProgress = TRUE)
# saveRDS(gwas_sumstats_unfiltered_AA, file = paste0(OUT_loc, "/gwas_sumstats_unfiltered.AA.rds"))
# gwas_sumstats_racer_unfiltered_AA <- subset(gwas_sumstats_unfiltered_AA,
# select = c("MarkerName", "rsID", "Chr", "Position", "Pvalue"))
#
# saveRDS(gwas_sumstats_racer_unfiltered_AA, file = paste0(OUT_loc, "/gwas_sumstats_unfiltered_racer.AA.rds"))
#
# gwas_sumstats_racer_unfiltered_AA <- readRDS(file = paste0(OUT_loc, "/gwas_sumstats_unfiltered_racer.AA.rds"))
#
# gwas_sumstats_AA <- fread(paste0(GWAS_loc,"/CAC1000G_AA_FINAL_FUMA.filtered.txt.gz"),
# showProgress = TRUE)
# saveRDS(gwas_sumstats_AA, file = paste0(OUT_loc, "/gwas_sumstats.AA.rds"))
#
# gwas_sumstats_racer_AA <- subset(gwas_sumstats_AA,
# select = c("MarkerName", "rsID", "Chr", "Position", "Pvalue"))
#
# saveRDS(gwas_sumstats_racer_AA, file = paste0(OUT_loc, "/gwas_sumstats_racer.AA.rds"))
# rm(gwas_sumstats_AA)
gwas_sumstats_racer_AA <- readRDS(file = paste0(OUT_loc, "/gwas_sumstats_racer.AA.rds"))Regional association plotting: EU-AA-ancestry
Top 11 loci
We are interested in 11 top loci. We will plot these using the EU-AA-ancestry data.
library(openxlsx)
variant_list <- read.xlsx(paste0(TARGET_loc, "/Variants.xlsx"), sheet = "TopLoci")
DT::datatable(variant_list)NAAll loci
Let’s do some plotting.
library(RACER)
# Make directory for plots
ifelse(!dir.exists(file.path(PROJECT_loc, "/RACER")),
dir.create(file.path(PROJECT_loc, "/RACER")),
FALSE)[1] FALSERACER_loc = paste0(PROJECT_loc,"/RACER")
variants_of_interest <- c(variant_list$rsID)
variants_of_interest_fewgenes <- c("rs9349379")#, "rs3844006", "rs2854746", "rs4977575", "rs9633535", "rs11063120", "rs9515203", "rs7182103")
for(VARIANT in variants_of_interest){
cat(paste0("Getting data for ", VARIANT,".\n"))
tempCHR <- subset(variant_list, rsID == VARIANT)[,5]
tempSTART <- subset(variant_list, rsID == VARIANT)[,17]
tempEND <- subset(variant_list, rsID == VARIANT)[,18]
tempVARIANTnr <- subset(variant_list, rsID == VARIANT)[,1]
cat("\nSubset required data.\n")
temp <- subset(gwas_sumstats_racer_EA_AA, Chr == tempCHR & (Position >= tempSTART & Position <= tempEND))
cat("\nFormatting association data.\n")
temp_f = RACER::formatRACER(assoc_data = temp, chr_col = 3, pos_col = 4, p_col = 5)
cat("\nGetting LD data.\n")
temp_f_ld = RACER::ldRACER(assoc_data = temp_f, rs_col = 2, pops = "EUR", lead_snp = VARIANT)
cat(paste0("\nPlotting region surrounding ", VARIANT," on ",tempCHR,":",tempSTART,"-",tempEND,".\n"))
p1 <- singlePlotRACER2(assoc_data = temp_f_ld,
chr = tempCHR, build = "hg19",
plotby = "snp", snp_plot = VARIANT,
label_lead = TRUE, gene_track_h = 2, gene_name_s = 1.75)
print(p1)
cat(paste0("Saving image for ", VARIANT,".\n"))
# ggsave(filename = paste0(RACER_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.png"), plot = last_plot())
# ggsave(filename = paste0(RACER_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.pdf"), plot = last_plot())
# ggsave(filename = paste0(RACER_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.eps"), plot = last_plot())
# rm(temp, p1,
# temp_f, temp_f_ld,
# tempCHR, tempSTART, tempEND,
# VARIANT, tempVARIANTnr)
}Getting data for rs9349379.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Getting LD data.All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs9349379...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs9349379&pop=EUR&r2_d=r2&token=c0f613f149ab"
[45%] Downloaded 32210 bytes...
[100%] Downloaded 70967 bytes...
Merging input association data with LD...
Plotting region surrounding rs9349379 on 6:12403957-13403957.Plotting by...
snp rs9349379
Reading in association data
Determining lead SNP
Generating PlotSaving image for rs9349379.
Getting data for rs3844006.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Getting LD data.All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs3844006...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs3844006&pop=EUR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 31016 bytes...
Merging input association data with LD...
Plotting region surrounding rs3844006 on 6:131595002-132595002.Plotting by...
snp rs3844006
Reading in association data
Determining lead SNP
Generating PlotSaving image for rs3844006.
Getting data for rs2854746.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Getting LD data.All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs2854746...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs2854746&pop=EUR&r2_d=r2&token=c0f613f149ab"
[44%] Downloaded 49152 bytes...
[59%] Downloaded 65518 bytes...
[74%] Downloaded 81920 bytes...
[89%] Downloaded 98286 bytes...
[100%] Downloaded 109298 bytes...
Merging input association data with LD...
Plotting region surrounding rs2854746 on 7:45460645-46460645.Plotting by...
snp rs2854746
Reading in association data
Determining lead SNP
Generating PlotSaving image for rs2854746.
Getting data for rs4977575.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Getting LD data.All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs4977575...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs4977575&pop=EUR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 26000 bytes...
Merging input association data with LD...
Plotting region surrounding rs4977575 on 9:21624744-22624744.Plotting by...
snp rs4977575
Reading in association data
Determining lead SNP
Generating PlotSaving image for rs4977575.
Getting data for rs10899970.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Getting LD data.All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs10899970...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs10899970&pop=EUR&r2_d=r2&token=c0f613f149ab"
[35%] Downloaded 57344 bytes...
[56%] Downloaded 90112 bytes...
[100%] Downloaded 160476 bytes...
Merging input association data with LD...
Plotting region surrounding rs10899970 on 10:44015716-45334720.Plotting by...
snp rs10899970
Reading in association data
Determining lead SNP
Generating PlotSaving image for rs10899970.
Getting data for rs9633535.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Getting LD data.All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs9633535...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs9633535&pop=EUR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 49363 bytes...
Merging input association data with LD...
Plotting region surrounding rs9633535 on 10:63336088-64336088.Plotting by...
snp rs9633535
Reading in association data
Determining lead SNP
Generating PlotSaving image for rs9633535.
Getting data for rs10762577.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Getting LD data.All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs10762577...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs10762577&pop=EUR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 92997 bytes...
Merging input association data with LD...
Plotting region surrounding rs10762577 on 10:75417431-76417431.Plotting by...
snp rs10762577
Reading in association data
Determining lead SNP
Generating PlotSaving image for rs10762577.
Getting data for rs11063120.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Getting LD data.All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs11063120...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs11063120&pop=EUR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 41883 bytes...
Merging input association data with LD...
Plotting region surrounding rs11063120 on 12:3986618-4986618.Plotting by...
snp rs11063120
Reading in association data
Determining lead SNP
Generating PlotSaving image for rs11063120.
Getting data for rs9515203.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Getting LD data.All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs9515203...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs9515203&pop=EUR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 30139 bytes...
Merging input association data with LD...
Plotting region surrounding rs9515203 on 13:110549623-111549623.Plotting by...
snp rs9515203
Reading in association data
Determining lead SNP
Generating PlotSaving image for rs9515203.
Getting data for rs7182103.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Getting LD data.All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs7182103...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs7182103&pop=EUR&r2_d=r2&token=c0f613f149ab"
[27%] Downloaded 32209 bytes...
[76%] Downloaded 90112 bytes...
[100%] Downloaded 118299 bytes...
Merging input association data with LD...
Plotting region surrounding rs7182103 on 15:78623946-79623946.Plotting by...
snp rs7182103
Reading in association data
Determining lead SNP
Generating PlotSaving image for rs7182103.
Getting data for rs7412.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Getting LD data.All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs7412...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs7412&pop=EUR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 38865 bytes...
Merging input association data with LD...
Plotting region surrounding rs7412 on 19:44912079-45912079.Plotting by...
snp rs7412
Reading in association data
Determining lead SNP
Generating PlotSaving image for rs7412.
Loci with many genes
These are genetic loci with many genes.
variants_of_interest_manygenes <- c("rs7412", "rs10762577")
for(VARIANT in variants_of_interest_manygenes){
cat(paste0("Getting data for ", VARIANT,".\n"))
tempCHR <- subset(variant_list, rsID == VARIANT)[,5]
tempSTART <- subset(variant_list, rsID == VARIANT)[,17]
tempEND <- subset(variant_list, rsID == VARIANT)[,18]
tempVARIANTnr <- subset(variant_list, rsID == VARIANT)[,1]
cat("\nSubset required data.\n")
temp <- subset(gwas_sumstats_racer_EA_AA, Chr == tempCHR & (Position >= tempSTART & Position <= tempEND))
cat("\nFormatting association data.\n")
temp_f = RACER::formatRACER(assoc_data = temp, chr_col = 3, pos_col = 4, p_col = 5)
cat("\nGetting LD data.\n")
temp_f_ld = RACER::ldRACER(assoc_data = temp_f, rs_col = 2, pops = "EUR", lead_snp = VARIANT)
cat(paste0("\nPlotting region surrounding ", VARIANT," on ",tempCHR,":",tempSTART,"-",tempEND,".\n"))
p1 <- singlePlotRACER2(assoc_data = temp_f_ld,
chr = tempCHR, build = "hg19",
plotby = "snp", snp_plot = VARIANT,
label_lead = TRUE, gene_track_h = 0.75, gene_name_s = 1.75)
print(p1)
cat(paste0("Saving image for ", VARIANT,".\n"))
# ggsave(filename = paste0(RACER_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.png"), plot = last_plot())
# ggsave(filename = paste0(RACER_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.pdf"), plot = last_plot())
# ggsave(filename = paste0(RACER_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.eps"), plot = last_plot())
rm(temp, p1,
temp_f, temp_f_ld,
tempCHR, tempSTART, tempEND,
VARIANT, tempVARIANTnr)
}Getting data for rs7412.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Getting LD data.All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs7412...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs7412&pop=EUR&r2_d=r2&token=c0f613f149ab"Merging input association data with LD...
Plotting region surrounding rs7412 on 19:44912079-45912079.Plotting by...
snp rs7412
Reading in association data
Determining lead SNP
Generating PlotSaving image for rs7412.
Getting data for rs10762577.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Getting LD data.All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs10762577...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs10762577&pop=EUR&r2_d=r2&token=c0f613f149ab"
[52%] Downloaded 48576 bytes...
[59%] Downloaded 55496 bytes...
[77%] Downloaded 71862 bytes...
[100%] Downloaded 92997 bytes...
Merging input association data with LD...
Plotting region surrounding rs10762577 on 10:75417431-76417431.Plotting by...
snp rs10762577
Reading in association data
Determining lead SNP
Generating PlotSaving image for rs10762577.
CXCL12
The CXCL12 genetic locus.
variants_of_interest_cxcl12 <- c("rs10899970")
for(VARIANT in variants_of_interest_cxcl12){
cat(paste0("Getting data for ", VARIANT,".\n"))
tempCHR <- subset(variant_list, rsID == VARIANT)[,5]
tempSTART <- subset(variant_list, rsID == VARIANT)[,17]
tempEND <- subset(variant_list, rsID == VARIANT)[,18]
tempVARIANTnr <- subset(variant_list, rsID == VARIANT)[,1]
cat("\nSubset required data.\n")
temp <- subset(gwas_sumstats_racer_EA_AA, Chr == tempCHR & (Position >= tempSTART & Position <= tempEND))
cat("\nFormatting association data.\n")
temp_f = RACER::formatRACER(assoc_data = temp, chr_col = 3, pos_col = 4, p_col = 5)
cat("\nGetting LD data.\n")
temp_f_ld = RACER::ldRACER(assoc_data = temp_f, rs_col = 2, pops = "EUR", lead_snp = VARIANT)
cat(paste0("\nPlotting region surrounding ", VARIANT," on ",tempCHR,":",tempSTART,"-",tempEND,".\n"))
p1 <- singlePlotRACER2(assoc_data = temp_f_ld,
chr = tempCHR, build = "hg19", set = "all",
plotby = "snp", snp_plot = VARIANT,
label_lead = TRUE, gene_track_h = 0.75, gene_name_s = 1.75)
print(p1)
cat(paste0("Saving image for ", VARIANT,".\n"))
# ggsave(filename = paste0(RACER_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.png"), plot = last_plot())
# ggsave(filename = paste0(RACER_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.pdf"), plot = last_plot())
# ggsave(filename = paste0(RACER_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.eps"), plot = last_plot())
rm(temp, p1,
temp_f, temp_f_ld,
tempCHR, tempSTART, tempEND,
VARIANT, tempVARIANTnr)
}Getting data for rs10899970.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Getting LD data.All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs10899970...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs10899970&pop=EUR&r2_d=r2&token=c0f613f149ab"
[45%] Downloaded 73710 bytes...
[86%] Downloaded 139246 bytes...
[100%] Downloaded 160476 bytes...
Merging input association data with LD...
Plotting region surrounding rs10899970 on 10:44015716-45334720.Plotting by...
snp rs10899970
Reading in association data
Determining lead SNP
Generating PlotSaving image for rs10899970.
Additional regional plots
Listing regions of interest
We want to create some regional association plots to combine with teh UCSC browser tracks, thus we need the exact same regions.
library(openxlsx)
add_list <- read.xlsx(paste0(TARGET_loc, "/Variants.xlsx"), sheet = "AdditionalPlots")
DT::datatable(add_list)NACredible Sets
We want to color the credible sets, which we load here.
credset <- as_tibble(fread(paste0(PROJECT_loc, "/CredibleSets/CAC_EUR_AFR_cred_set_all_loci_50kb.txt")))
credsetCombining GWAS with Credible Set
We want to add the posterior probabilities and make a variable to color by.
gwas_sumstats_racer_credset <- merge(gwas_sumstats_racer_EA_AA,
credset %>% select(RSID, Posterior_Prob),
sort = FALSE,
by.x = "rsID", by.y = "RSID", all.x = TRUE) %>%
# mutate(., Posterior_Prob = ifelse(is.na(Posterior_Prob), 0, Posterior_Prob)) %>%
mutate(CredSet = case_when(Posterior_Prob > 0 ~ '95% credible set',
TRUE ~ 'not in credible set'))
head(gwas_sumstats_racer_credset)
table(gwas_sumstats_racer_credset$CredSet)
95% credible set not in credible set
103 8585944 summary(gwas_sumstats_racer_credset$Posterior_Prob) Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
0 0 0 0 0 1 8585944 Plotting
library(RACER)
# library(plotly)
# Make directory for plots
ifelse(!dir.exists(file.path(PROJECT_loc, "/RACER")),
dir.create(file.path(PROJECT_loc, "/RACER")),
FALSE)[1] FALSERACER_loc = paste0(PROJECT_loc,"/RACER")
variants_of_interest <- c(add_list$rsID)
for(VARIANT in variants_of_interest){
cat(paste0("Getting data for ", VARIANT,".\n"))
tempCHR <- subset(add_list, rsID == VARIANT)[,4]
tempSTART <- subset(add_list, rsID == VARIANT)[,5]
tempEND <- subset(add_list, rsID == VARIANT)[,6]
tempNAME <- subset(add_list, rsID == VARIANT)[,3]
cat("\nSubset required data.\n")
temp <- subset(gwas_sumstats_racer_credset, Chr == tempCHR & (Position >= tempSTART & Position <= tempEND))
cat("\nFormatting association data.\n")
temp_f = RACER::formatRACER(assoc_data = temp, chr_col = 3, pos_col = 4, p_col = 5)
cat("\nGetting LD data.\n")
# temp_f_ld = RACER::ldRACER(assoc_data = temp_f, rs_col = 2, pops = "EUR", lead_snp = VARIANT)
cat(paste0("\nPlotting region surrounding ", VARIANT," on ",tempCHR,":",tempSTART,"-",tempEND,".\n"))
p1 <- singlePlotRACER2(assoc_data = temp_f,
chr = tempCHR, build = "hg19",
plotby = "coord", snp_plot = VARIANT,
start_plot = tempSTART, end_plot = tempEND,
label_lead = FALSE,
grey_colors = FALSE,
cred_set = TRUE,
gene_track_h = 3, gene_name_s = 1.75)
print(p1)
cat(paste0("Saving image for ", VARIANT,".\n"))
# ggsave(filename = paste0(RACER_loc, "/", tempNAME, ".", Today, ".",VARIANT,".",tempSTART,".",tempEND,".regional_assoc.png"), plot = p1)
# ggsave(filename = paste0(RACER_loc, "/", tempNAME, ".", Today, ".",VARIANT,".",tempSTART,".",tempEND,".regional_assoc.pdf"), plot = p1)
# ggsave(filename = paste0(RACER_loc, "/", tempNAME, ".", Today, ".",VARIANT,".",tempSTART,".",tempEND,".regional_assoc.eps"), plot = p1)
# print(ggplotly(p1))
rm(temp, p1,
temp_f,
tempCHR, tempSTART, tempEND,
VARIANT, tempNAME)
}Getting data for rs9633535.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Getting LD data.
Plotting region surrounding rs9633535 on 10:63584853-63921073.Association Data Set is missing LD data, the resulting plot won't have LD information, but you can add it using the ldRACER.R function.
Plotting by...
coord
Reading in association data
Collecting posterior probabilities
Generating Plot
Scale for 'colour' is already present. Adding another scale for 'colour', which will replace the existing scale.Saving image for rs9633535.
Getting data for rs2854746.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Getting LD data.
Plotting region surrounding rs2854746 on 7:45894617-46054070.Association Data Set is missing LD data, the resulting plot won't have LD information, but you can add it using the ldRACER.R function.
Plotting by...
coord
Reading in association data
Collecting posterior probabilities
Generating Plot
Scale for 'colour' is already present. Adding another scale for 'colour', which will replace the existing scale.Saving image for rs2854746.
Getting data for rs3844006.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Getting LD data.
Plotting region surrounding rs3844006 on 6:131937915-132289374.Association Data Set is missing LD data, the resulting plot won't have LD information, but you can add it using the ldRACER.R function.
Plotting by...
coord
Reading in association data
Collecting posterior probabilities
Generating Plot
Scale for 'colour' is already present. Adding another scale for 'colour', which will replace the existing scale.Saving image for rs3844006.
Regional association plots in African-American
Note here that we plot the region, and not based on the lead variant of the EA-AA analyses.
library(RACER)
# Make directory for plots
ifelse(!dir.exists(file.path(PROJECT_loc, "/RACER_AA")),
dir.create(file.path(PROJECT_loc, "/RACER_AA")),
FALSE)[1] FALSERACER_AA_loc = paste0(PROJECT_loc,"/RACER_AA")
variants_of_interest <- c(variant_list$rsID)
variants_of_interest_fewgenes <- c("rs9349379",
"rs3844006", # throws an error which I don't understand immediately - could be that the variant is not present in AA 1000G data
"rs2854746", "rs4977575",
"rs10899970",
"rs9633535", # this one throws an LD error
"rs10762577",
"rs11063120", "rs9515203", "rs7182103",
"rs7412")
for(VARIANT in variants_of_interest){
cat(paste0("Getting data for ", VARIANT,".\n"))
tempCHR <- subset(variant_list, rsID == VARIANT)[,5]
tempSTART <- subset(variant_list, rsID == VARIANT)[,17]
tempEND <- subset(variant_list, rsID == VARIANT)[,18]
tempVARIANTnr <- subset(variant_list, rsID == VARIANT)[,1]
cat("\nSubset required data.\n")
temp <- subset(gwas_sumstats_racer_AA, Chr == tempCHR & (Position >= tempSTART & Position <= tempEND))
cat("\nFormatting association data.\n")
temp_f = RACER::formatRACER(assoc_data = temp, chr_col = 3, pos_col = 4, p_col = 5)
# cat("\nGetting LD data.\n")
temp_f_ld = RACER::ldRACER(assoc_data = temp_f, rs_col = 2, pops = "AFR", lead_snp = VARIANT)
cat(paste0("\nPlotting region surrounding ", VARIANT," on ",tempCHR,":",tempSTART,"-",tempEND,".\n"))
p1 <- singlePlotRACER2(assoc_data = temp_f,
chr = tempCHR, build = "hg19",
plotby = "coord", snp_plot = VARIANT,
start_plot = tempSTART, end_plot = tempEND,
label_lead = FALSE, gene_track_h = 2, gene_name_s = 1.75)
print(p1)
cat(paste0("Saving image for ", VARIANT,".\n"))
ggsave(filename = paste0(RACER_AA_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.AA.png"), plot = last_plot())
ggsave(filename = paste0(RACER_AA_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.AA.pdf"), plot = last_plot())
ggsave(filename = paste0(RACER_AA_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.AA.eps"), plot = last_plot())
rm(temp, p1,
temp_f, temp_f_ld,
tempCHR, tempSTART, tempEND,
VARIANT, tempVARIANTnr)
}Getting data for rs9349379.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs9349379...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs9349379&pop=AFR&r2_d=r2&token=c0f613f149ab"
[16%] Downloaded 15980 bytes...
[33%] Downloaded 32210 bytes...
[50%] Downloaded 48576 bytes...
[100%] Downloaded 95751 bytes...
Merging input association data with LD...
Plotting region surrounding rs9349379 on 6:12403957-13403957.Association Data Set is missing LD data, the resulting plot won't have LD information, but you can add it using the ldRACER.R function.
Plotting by...
coord
Reading in association data
Generating PlotSaving image for rs9349379.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageGetting data for rs3844006.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs3844006...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs3844006&pop=AFR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 72560 bytes...
Merging input association data with LD...
Plotting region surrounding rs3844006 on 6:131595002-132595002.Association Data Set is missing LD data, the resulting plot won't have LD information, but you can add it using the ldRACER.R function.
Plotting by...
coord
Reading in association data
Generating PlotSaving image for rs3844006.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageGetting data for rs2854746.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs2854746...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs2854746&pop=AFR&r2_d=r2&token=c0f613f149ab"
[22%] Downloaded 15980 bytes...
[100%] Downloaded 71651 bytes...
Merging input association data with LD...
Plotting region surrounding rs2854746 on 7:45460645-46460645.Association Data Set is missing LD data, the resulting plot won't have LD information, but you can add it using the ldRACER.R function.
Plotting by...
coord
Reading in association data
Generating PlotSaving image for rs2854746.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageGetting data for rs4977575.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs4977575...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs4977575&pop=AFR&r2_d=r2&token=c0f613f149ab"
[17%] Downloaded 15980 bytes...
[100%] Downloaded 93249 bytes...
Merging input association data with LD...
Plotting region surrounding rs4977575 on 9:21624744-22624744.Association Data Set is missing LD data, the resulting plot won't have LD information, but you can add it using the ldRACER.R function.
Plotting by...
coord
Reading in association data
Generating PlotSaving image for rs4977575.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageGetting data for rs10899970.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs10899970...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs10899970&pop=AFR&r2_d=r2&token=c0f613f149ab"
[35%] Downloaded 64977 bytes...
[61%] Downloaded 114688 bytes...
[70%] Downloaded 131054 bytes...
[79%] Downloaded 147456 bytes...
[100%] Downloaded 185291 bytes...
Merging input association data with LD...
Plotting region surrounding rs10899970 on 10:44015716-45334720.Association Data Set is missing LD data, the resulting plot won't have LD information, but you can add it using the ldRACER.R function.
Plotting by...
coord
Reading in association data
Generating PlotSaving image for rs10899970.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageGetting data for rs9633535.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs9633535...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs9633535&pop=AFR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 42221 bytes...
Merging input association data with LD...
Plotting region surrounding rs9633535 on 10:63336088-64336088.Association Data Set is missing LD data, the resulting plot won't have LD information, but you can add it using the ldRACER.R function.
Plotting by...
coord
Reading in association data
Generating PlotSaving image for rs9633535.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageGetting data for rs10762577.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs10762577...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs10762577&pop=AFR&r2_d=r2&token=c0f613f149ab"
[78%] Downloaded 114111 bytes...
[90%] Downloaded 130513 bytes...
[100%] Downloaded 144558 bytes...
Merging input association data with LD...
Plotting region surrounding rs10762577 on 10:75417431-76417431.Association Data Set is missing LD data, the resulting plot won't have LD information, but you can add it using the ldRACER.R function.
Plotting by...
coord
Reading in association data
Generating PlotSaving image for rs10762577.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageGetting data for rs11063120.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs11063120...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs11063120&pop=AFR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 46036 bytes...
Merging input association data with LD...
Plotting region surrounding rs11063120 on 12:3986618-4986618.Association Data Set is missing LD data, the resulting plot won't have LD information, but you can add it using the ldRACER.R function.
Plotting by...
coord
Reading in association data
Generating PlotSaving image for rs11063120.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageGetting data for rs9515203.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs9515203...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs9515203&pop=AFR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 41921 bytes...
Merging input association data with LD...
Plotting region surrounding rs9515203 on 13:110549623-111549623.Association Data Set is missing LD data, the resulting plot won't have LD information, but you can add it using the ldRACER.R function.
Plotting by...
coord
Reading in association data
Generating PlotSaving image for rs9515203.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageGetting data for rs7182103.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs7182103...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs7182103&pop=AFR&r2_d=r2&token=c0f613f149ab"
[72%] Downloaded 81343 bytes...
[87%] Downloaded 97745 bytes...
[100%] Downloaded 111697 bytes...
Merging input association data with LD...
Plotting region surrounding rs7182103 on 15:78623946-79623946.Association Data Set is missing LD data, the resulting plot won't have LD information, but you can add it using the ldRACER.R function.
Plotting by...
coord
Reading in association data
Generating PlotSaving image for rs7182103.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageGetting data for rs7412.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs7412...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs7412&pop=AFR&r2_d=r2&token=c0f613f149ab"
[48%] Downloaded 15980 bytes...
[100%] Downloaded 33190 bytes...
Merging input association data with LD...
Plotting region surrounding rs7412 on 19:44912079-45912079.Association Data Set is missing LD data, the resulting plot won't have LD information, but you can add it using the ldRACER.R function.
Plotting by...
coord
Reading in association data
Generating PlotSaving image for rs7412.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Approximate Bayes Factor colocalisation analyses
The idea behind the Approximate Bayes Factor (ABF) analysis is that the association of each trait with SNPs in a region may be summarised by a vector of 0s and at most a single 1, with the 1 indicating the causal SNP (so, assuming a single causal SNP for each trait).
The posterior probability of each possible configuration can be calculated and so, crucially, can the posterior probabilities that the traits share their configurations. This allows us to estimate the support for the following cases, i.e. hypotheses:
𝐻0: neither trait has a genetic association in the region 𝐻1: only trait 1 has a genetic association in the region 𝐻2: only trait 2 has a genetic association in the region 𝐻3: both traits are associated, but with different causal variants 𝐻4: both traits are associated and share a single causal variant
To what extent do the loci between European and African-American ancestries overlap? We are working on the assumption that the 11 loci are the loci and will test whether these overlap, i.e. colocalize.
Preparation
Let’s make sure we have remotes and coloc
installed.
if(!require("remotes"))
install.packages("remotes") # if necessary
library(remotes)
install_github("chr1swallace/coloc@main",
build_vignettes = TRUE)Using github PAT from envvar GITHUB_PAT
Skipping install of 'coloc' from a github remote, the SHA1 (0b0701a5) has not changed since last install.
Use `force = TRUE` to force installationlibrary(coloc)This is coloc version 5.2.0Visualization
We can create mirror and scatter plot for each region.
library(RACER)
# Make directory for plots
ifelse(!dir.exists(file.path(PROJECT_loc, "/RACER_EA_vs_AA")),
dir.create(file.path(PROJECT_loc, "/RACER_EA_vs_AA")),
FALSE)[1] FALSERACER_EA_vs_AA_loc = paste0(PROJECT_loc,"/RACER_EA_vs_AA")
variants_of_interest <- c(variant_list$rsID)
# variants_of_interest_fewgenes <- c("rs9349379",
# "rs3844006", # throws an error which I don't understand immediately - could be that the variant is not present in AA 1000G data
# "rs2854746", "rs4977575",
# "rs10899970",
# "rs9633535",
# "rs10762577",
# "rs11063120", "rs9515203", "rs7182103",
# "rs7412")
variants_of_interest_fewgenes <- c("rs10762577",
"rs11063120", "rs9515203", "rs7182103",
"rs7412")
for(VARIANT in variants_of_interest){
cat(paste0("Getting data for ", VARIANT,".\n"))
tempCHR <- subset(variant_list, rsID == VARIANT)[,5]
tempSTART <- subset(variant_list, rsID == VARIANT)[,17]
tempEND <- subset(variant_list, rsID == VARIANT)[,18]
tempVARIANTnr <- subset(variant_list, rsID == VARIANT)[,1]
cat("\nSubset required data.\n")
temp1 <- subset(gwas_sumstats_racer_EA, Chr == tempCHR & (Position >= tempSTART & Position <= tempEND))
temp2 <- subset(gwas_sumstats_racer_AA, Chr == tempCHR & (Position >= tempSTART & Position <= tempEND))
cat("\nFormatting association data.\n")
temp_f1 = RACER::formatRACER(assoc_data = temp1, chr_col = 3, pos_col = 4, p_col = 5)
temp_f2 = RACER::formatRACER(assoc_data = temp2, chr_col = 3, pos_col = 4, p_col = 5)
# cat("\nGetting LD data.\n")
temp_f_ld1 = RACER::ldRACER(assoc_data = temp_f1, rs_col = 2, pops = "EUR", lead_snp = VARIANT)
temp_f_ld2 = RACER::ldRACER(assoc_data = temp_f2, rs_col = 2, pops = "AFR", lead_snp = VARIANT)
cat(paste0("\nPlotting region surrounding ", VARIANT," on ",tempCHR,":",tempSTART,"-",tempEND,".\n"))
p1 <- mirrorPlotRACER(assoc_data1 = temp_f_ld1,
assoc_data2 = temp_f_ld2,
chr = tempCHR,
name1 = "European ancestry",
name2 = "African-American ancestry",
plotby = "coord",
start_plot = tempSTART, end_plot = tempEND)
print(p1)
p2 <- scatterPlotRACER(assoc_data1 = temp_f_ld1,
assoc_data2 = temp_f_ld2,
chr = tempCHR,
name1 = "European ancestry",
name2 = "African-American ancestry",
region_start = tempSTART,
region_end = tempEND,
ld_df = 1)
print(p2)
cat(paste0("Saving image for ", VARIANT,".\n"))
ggsave(filename = paste0(RACER_EA_vs_AA_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_mirror.EA_vs_AA.png"), plot = p1)
ggsave(filename = paste0(RACER_EA_vs_AA_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_mirror.EA_vs_AA.pdf"), plot = p1)
ggsave(filename = paste0(RACER_EA_vs_AA_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_mirror.EA_vs_AA.eps"), plot = p1)
cat(paste0("Saving image for ", VARIANT,".\n"))
ggsave(filename = paste0(RACER_EA_vs_AA_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_scatter.EA_vs_AA.png"), plot = p2)
ggsave(filename = paste0(RACER_EA_vs_AA_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_scatter.EA_vs_AA.pdf"), plot = p2)
ggsave(filename = paste0(RACER_EA_vs_AA_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_scatter.EA_vs_AA.eps"), plot = p2)
rm(temp1, temp2,
p1,p2,
temp_f1, temp_f_ld1,
temp_f2, temp_f_ld2,
tempCHR, tempSTART, tempEND,
VARIANT, tempVARIANTnr)
}Getting data for rs9349379.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Formating association data...
Processing -log10(p-values)...
Preparing association data...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs9349379...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs9349379&pop=EUR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 70967 bytes...
Merging input association data with LD...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs9349379...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs9349379&pop=AFR&r2_d=r2&token=c0f613f149ab"
[50%] Downloaded 48576 bytes...
[57%] Downloaded 55496 bytes...
[100%] Downloaded 95751 bytes...
Merging input association data with LD...
Plotting region surrounding rs9349379 on 6:12403957-13403957.Plotting by...
coord
Reading in association data
Generating plot.
Scale for 'y' is already present. Adding another scale for 'y', which will replace the existing scale.
Scale for 'y' is already present. Adding another scale for 'y', which will replace the existing scale.
Reading in association data
Generating plot.Saving image for rs9349379.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageSaving image for rs9349379.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageGetting data for rs3844006.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Formating association data...
Processing -log10(p-values)...
Preparing association data...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs3844006...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs3844006&pop=EUR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 31016 bytes...
Merging input association data with LD...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs3844006...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs3844006&pop=AFR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 72560 bytes...
Merging input association data with LD...
Plotting region surrounding rs3844006 on 6:131595002-132595002.Plotting by...
coord
Reading in association data
Generating plot.
Scale for 'y' is already present. Adding another scale for 'y', which will replace the existing scale.
Scale for 'y' is already present. Adding another scale for 'y', which will replace the existing scale.
Reading in association data
Generating plot.Saving image for rs3844006.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageSaving image for rs3844006.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageGetting data for rs2854746.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Formating association data...
Processing -log10(p-values)...
Preparing association data...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs2854746...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs2854746&pop=EUR&r2_d=r2&token=c0f613f149ab"
[80%] Downloaded 88263 bytes...
[95%] Downloaded 104629 bytes...
[100%] Downloaded 109298 bytes...
Merging input association data with LD...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs2854746...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs2854746&pop=AFR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 71651 bytes...
Merging input association data with LD...
Plotting region surrounding rs2854746 on 7:45460645-46460645.Plotting by...
coord
Reading in association data
Generating plot.
Scale for 'y' is already present. Adding another scale for 'y', which will replace the existing scale.
Scale for 'y' is already present. Adding another scale for 'y', which will replace the existing scale.
Reading in association data
Generating plot.Saving image for rs2854746.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageSaving image for rs2854746.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageGetting data for rs4977575.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Formating association data...
Processing -log10(p-values)...
Preparing association data...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs4977575...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs4977575&pop=EUR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 26000 bytes...
Merging input association data with LD...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs4977575...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs4977575&pop=AFR&r2_d=r2&token=c0f613f149ab"
[59%] Downloaded 55496 bytes...
[94%] Downloaded 88264 bytes...
[100%] Downloaded 93249 bytes...
Merging input association data with LD...
Plotting region surrounding rs4977575 on 9:21624744-22624744.Plotting by...
coord
Reading in association data
Generating plot.
Scale for 'y' is already present. Adding another scale for 'y', which will replace the existing scale.
Scale for 'y' is already present. Adding another scale for 'y', which will replace the existing scale.
Reading in association data
Generating plot.Saving image for rs4977575.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageSaving image for rs4977575.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageGetting data for rs10899970.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Formating association data...
Processing -log10(p-values)...
Preparing association data...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs10899970...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs10899970&pop=EUR&r2_d=r2&token=c0f613f149ab"
[51%] Downloaded 81920 bytes...
[61%] Downloaded 98286 bytes...
[71%] Downloaded 114688 bytes...
[81%] Downloaded 131054 bytes...
[100%] Downloaded 160476 bytes...
Merging input association data with LD...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs10899970...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs10899970&pop=AFR&r2_d=r2&token=c0f613f149ab"
[8%] Downloaded 15979 bytes...
[30%] Downloaded 57344 bytes...
[39%] Downloaded 73710 bytes...
[75%] Downloaded 139246 bytes...
[84%] Downloaded 155648 bytes...
[100%] Downloaded 185291 bytes...
Merging input association data with LD...
Plotting region surrounding rs10899970 on 10:44015716-45334720.Plotting by...
coord
Reading in association data
Generating plot.
Scale for 'y' is already present. Adding another scale for 'y', which will replace the existing scale.
Scale for 'y' is already present. Adding another scale for 'y', which will replace the existing scale.
Reading in association data
Generating plot.Saving image for rs10899970.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageSaving image for rs10899970.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageGetting data for rs9633535.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Formating association data...
Processing -log10(p-values)...
Preparing association data...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs9633535...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs9633535&pop=EUR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 49363 bytes...
Merging input association data with LD...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs9633535...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs9633535&pop=AFR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 42221 bytes...
Merging input association data with LD...
Plotting region surrounding rs9633535 on 10:63336088-64336088.Plotting by...
coord
Reading in association data
Generating plot.
Scale for 'y' is already present. Adding another scale for 'y', which will replace the existing scale.
Scale for 'y' is already present. Adding another scale for 'y', which will replace the existing scale.
Reading in association data
Generating plot.Saving image for rs9633535.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageSaving image for rs9633535.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageGetting data for rs10762577.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Formating association data...
Processing -log10(p-values)...
Preparing association data...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs10762577...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs10762577&pop=EUR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 92997 bytes...
Merging input association data with LD...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs10762577...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs10762577&pop=AFR&r2_d=r2&token=c0f613f149ab"
[56%] Downloaded 81920 bytes...
[100%] Downloaded 144558 bytes...
Merging input association data with LD...
Plotting region surrounding rs10762577 on 10:75417431-76417431.Plotting by...
coord
Reading in association data
Generating plot.
Scale for 'y' is already present. Adding another scale for 'y', which will replace the existing scale.
Scale for 'y' is already present. Adding another scale for 'y', which will replace the existing scale.
Reading in association data
Generating plot.Saving image for rs10762577.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageSaving image for rs10762577.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageGetting data for rs11063120.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Formating association data...
Processing -log10(p-values)...
Preparing association data...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs11063120...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs11063120&pop=EUR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 41883 bytes...
Merging input association data with LD...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs11063120...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs11063120&pop=AFR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 46036 bytes...
Merging input association data with LD...
Plotting region surrounding rs11063120 on 12:3986618-4986618.Plotting by...
coord
Reading in association data
Generating plot.
Scale for 'y' is already present. Adding another scale for 'y', which will replace the existing scale.
Scale for 'y' is already present. Adding another scale for 'y', which will replace the existing scale.
Reading in association data
Generating plot.Saving image for rs11063120.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageSaving image for rs11063120.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageGetting data for rs9515203.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Formating association data...
Processing -log10(p-values)...
Preparing association data...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs9515203...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs9515203&pop=EUR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 30139 bytes...
Merging input association data with LD...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs9515203...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs9515203&pop=AFR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 41921 bytes...
Merging input association data with LD...
Plotting region surrounding rs9515203 on 13:110549623-111549623.Plotting by...
coord
Reading in association data
Generating plot.
Scale for 'y' is already present. Adding another scale for 'y', which will replace the existing scale.
Scale for 'y' is already present. Adding another scale for 'y', which will replace the existing scale.
Reading in association data
Generating plot.Saving image for rs9515203.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageSaving image for rs9515203.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageGetting data for rs7182103.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Formating association data...
Processing -log10(p-values)...
Preparing association data...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs7182103...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs7182103&pop=EUR&r2_d=r2&token=c0f613f149ab"
[55%] Downloaded 65518 bytes...
[83%] Downloaded 98286 bytes...
[100%] Downloaded 118299 bytes...
Merging input association data with LD...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs7182103...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs7182103&pop=AFR&r2_d=r2&token=c0f613f149ab"
[64%] Downloaded 71861 bytes...
[79%] Downloaded 88263 bytes...
[100%] Downloaded 111697 bytes...
Merging input association data with LD...
Plotting region surrounding rs7182103 on 15:78623946-79623946.Plotting by...
coord
Reading in association data
Generating plot.
Scale for 'y' is already present. Adding another scale for 'y', which will replace the existing scale.
Scale for 'y' is already present. Adding another scale for 'y', which will replace the existing scale.
Reading in association data
Generating plot.Saving image for rs7182103.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageSaving image for rs7182103.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageGetting data for rs7412.
Subset required data.
Formatting association data.Formating association data...
Processing -log10(p-values)...
Preparing association data...
Formating association data...
Processing -log10(p-values)...
Preparing association data...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs7412...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs7412&pop=EUR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 38865 bytes...
Merging input association data with LD...
All inputs are go!
Reading in association data...
Populations selected.
Calculating LD using rs7412...[1] "https://ldlink.nci.nih.gov/LDlinkRest/ldproxy?var=rs7412&pop=AFR&r2_d=r2&token=c0f613f149ab"
[100%] Downloaded 33190 bytes...
Merging input association data with LD...
Plotting region surrounding rs7412 on 19:44912079-45912079.Plotting by...
coord
Reading in association data
Generating plot.
Scale for 'y' is already present. Adding another scale for 'y', which will replace the existing scale.
Scale for 'y' is already present. Adding another scale for 'y', which will replace the existing scale.
Warning: Removed 1 row(s) containing missing values (geom_path).
Warning: Removed 1 rows containing missing values (geom_text).
Reading in association data
Generating plot.Saving image for rs7412.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in imageSaving image for rs7412.Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Saving 7.29 x 4.51 in image
Quantify colocalization
We want to quantify the overlap.
Setting up an output directory for coloc.
ifelse(!dir.exists(file.path(PROJECT_loc, "/COLOC_EA_vs_AA")),
dir.create(file.path(PROJECT_loc, "/COLOC_EA_vs_AA")),
FALSE)[1] FALSECOLOC_EA_vs_AA_loc = paste0(PROJECT_loc,"/COLOC_EA_vs_AA")Data colleaction
Preparing data for coloc: we require beta and standard
errors for coloc.
First the European ancestry data, next the African-American ancestry data.
gwas_sumstats_EA <- fread(paste0(GWAS_loc,"/CAC1000G_EA_FINAL_FUMA.txt.gz"),
showProgress = TRUE)|--------------------------------------------------|
|==================================================|names(gwas_sumstats_EA)[names(gwas_sumstats_EA) == "Pos"] <- "Position"
gwas_sumstats_racer_EA <- subset(gwas_sumstats_EA,
select = c("MarkerName", "rsID", "Chr", "Position",
"Effect", "StdErr",
"Pvalue",
"N"))
rm(gwas_sumstats_EA)gwas_sumstats_AA <- fread(paste0(GWAS_loc,"/CAC1000G_AA_FINAL_FUMA.filtered.txt.gz"),
showProgress = TRUE)|--------------------------------------------------|
|==================================================|names(gwas_sumstats_AA)[names(gwas_sumstats_AA) == "Pos"] <- "Position"
gwas_sumstats_racer_AA <- subset(gwas_sumstats_AA,
select = c("MarkerName", "rsID", "Chr", "Position",
"Effect", "StdErr",
"Pvalue",
"N"))Error in `[.data.table`(x, r, vars, with = FALSE) :
column(s) not found: StdErrColocalization
Now we are reading to formally test the colocalization per trait.
Note that trait 1 = European ancestry; trait 2
= African-American ancestry.
for(VARIANT in variants_of_interest){
cat(paste0("Getting data for ", VARIANT,".\n"))
tempCHR <- subset(variant_list, rsID == VARIANT)[,5]
tempSTART <- subset(variant_list, rsID == VARIANT)[,17]
tempEND <- subset(variant_list, rsID == VARIANT)[,18]
tempVARIANTnr <- subset(variant_list, rsID == VARIANT)[,1]
cat("\nSubset required data.\n")
temp1 <- subset(gwas_sumstats_racer_EA, Chr == tempCHR & (Position >= tempSTART & Position <= tempEND))
temp2 <- subset(gwas_sumstats_racer_AA, Chr == tempCHR & (Position >= tempSTART & Position <= tempEND))
cat("\nCheck temp1 data.\n")
temp1 <- rename_with(temp1, tolower)
# correcting column names
temp1 <- rename(temp1, beta = effect)
temp1 <- rename(temp1, se = stderr)
temp1 <- rename(temp1, snp = rsid)
# calculating things
temp1$varbeta <- temp1$se^2
temp1 <- as.list(temp1) # critical, as coloc expects a list of variables
temp1$type <- "quant"
temp1$sdY <- 1
coloc::check_dataset(temp1, warn.minp = 1e-10)
cat("\nCheck temp2 data.\n")
temp2 <- rename_with(temp2, tolower)
# correcting column names
temp2 <- rename(temp2, beta = effect)
temp2 <- rename(temp2, se = stderr)
temp2 <- rename(temp2, snp = rsid)
# calculating things
temp2$varbeta <- temp2$se^2
temp2 <- as.list(temp2) # critical, as coloc expects a list of variables
temp2$type <- "quant"
temp2$sdY <- 1
coloc::check_dataset(temp2, warn.minp = 1e-10)
cat("\nPlot required data.\n")
plot_dataset(temp1)
plot_dataset(temp2)
res_temp1_vs_temp2_single <- coloc.abf(dataset1 = temp1,
dataset2 = temp2)
cat("\nColocalization.\n")
print(res_temp1_vs_temp2_single)
write_lines(res_temp1_vs_temp2_single, file = paste0(COLOC_EA_vs_AA_loc, "/res_EA_vs_AA_single.",
tempVARIANTnr,".",tempCHR,"_",tempSTART,"_",tempEND,".txt"))
coloc::sensitivity(res_temp1_vs_temp2_single, "H4 > 0.9")
# Step 1: Call the pdf command to start the plot
pdf(file = paste0(COLOC_EA_vs_AA_loc, "/res_EA_vs_AA_single.",
tempVARIANTnr,".",tempCHR,"_",tempSTART,"_",tempEND,".pdf")) # The directory you want to save the file in
# Step 2: Create the plot with R code
coloc::sensitivity(res_temp1_vs_temp2_single, "H4 > 0.9")
# Step 3: Run dev.off() to create the file!
dev.off()
rm(temp1, temp2,
# p1,p2,
# temp_f1, temp_f_ld1,
# temp_f2, temp_f_ld2,
tempCHR, tempSTART, tempEND,
VARIANT, tempVARIANTnr)
}Getting data for rs9349379.
Subset required data.
Check temp1 data.
Check temp2 data.Warning in coloc::check_dataset(temp2, warn.minp = 1e-10) :
minimum p value is: 0.0010709
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.
Plot required data.
Warning in check_dataset(d = dataset2, 2) :
minimum p value is: 0.0010709
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.PP.H0.abf PP.H1.abf PP.H2.abf PP.H3.abf PP.H4.abf
2.64e-21 4.40e-01 1.86e-22 3.04e-02 5.30e-01
[1] "PP abf for shared variant: 53%"
Colocalization.Coloc analysis of trait 1, trait 2
SNP Priors p1 p2 p12
1e-04 1e-04 1e-05
Hypothesis Priors H0 H1 H2 H3 H4
0.7610836 0.1082 0.1082 0.01169642 0.01082
Posterior nsnps H0 H1 H2 H3 H4
1.082000e+03 2.643499e-21 4.401108e-01 1.857062e-22 3.038835e-02 5.295009e-01 Results fail decision rule H4 > 0.9
Results fail decision rule H4 > 0.9Getting data for rs3844006.
Subset required data.
Check temp1 data.Warning in coloc::check_dataset(temp1, warn.minp = 1e-10) :
minimum p value is: 3.5548e-06
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.
Check temp2 data.Warning in coloc::check_dataset(temp2, warn.minp = 1e-10) :
minimum p value is: 9.5528e-05
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.
Plot required data.
Warning in check_dataset(d = dataset1, 1) :
minimum p value is: 3.5548e-06
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.
Warning in check_dataset(d = dataset2, 2) :
minimum p value is: 9.5528e-05
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.PP.H0.abf PP.H1.abf PP.H2.abf PP.H3.abf PP.H4.abf
0.21300 0.06910 0.02150 0.00628 0.69000
[1] "PP abf for shared variant: 69%"
Colocalization.Coloc analysis of trait 1, trait 2
SNP Priors p1 p2 p12
1e-04 1e-04 1e-05
Hypothesis Priors H0 H1 H2 H3 H4
0.8049601 0.0891 0.0891 0.0079299 0.00891
Posterior nsnps H0 H1 H2 H3 H4
8.910000e+02 2.127262e-01 6.907437e-02 2.147238e-02 6.281855e-03 6.904452e-01 Results fail decision rule H4 > 0.9
Results fail decision rule H4 > 0.9Getting data for rs2854746.
Subset required data.
Check temp1 data.Warning in coloc::check_dataset(temp1, warn.minp = 1e-10) :
minimum p value is: 1.7731e-07
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.
Check temp2 data.Warning in coloc::check_dataset(temp2, warn.minp = 1e-10) :
minimum p value is: 0.00072736
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.
Plot required data.
Warning in check_dataset(d = dataset2, 2) :
minimum p value is: 0.00072736
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.PP.H0.abf PP.H1.abf PP.H2.abf PP.H3.abf PP.H4.abf
0.5990 0.1080 0.0686 0.0122 0.2120
[1] "PP abf for shared variant: 21.2%"
Colocalization.Coloc analysis of trait 1, trait 2
SNP Priors p1 p2 p12
1e-04 1e-04 1e-05
Hypothesis Priors H0 H1 H2 H3 H4
0.7459854 0.1147 0.1147 0.01314462 0.01147
Posterior nsnps H0 H1 H2 H3 H4
1.147000e+03 5.987696e-01 1.080300e-01 6.855817e-02 1.215678e-02 2.124855e-01 Results fail decision rule H4 > 0.9
Results fail decision rule H4 > 0.9Getting data for rs4977575.
Subset required data.
Check temp1 data.
Check temp2 data.Warning in coloc::check_dataset(temp2, warn.minp = 1e-10) :
minimum p value is: 0.00077724
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.
Plot required data.
Warning in check_dataset(d = dataset2, 2) :
minimum p value is: 0.00077724
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.PP.H0.abf PP.H1.abf PP.H2.abf PP.H3.abf PP.H4.abf
3.82e-38 4.04e-01 3.70e-39 3.86e-02 5.58e-01
[1] "PP abf for shared variant: 55.8%"
Colocalization.Coloc analysis of trait 1, trait 2
SNP Priors p1 p2 p12
1e-04 1e-04 1e-05
Hypothesis Priors H0 H1 H2 H3 H4
0.7394555 0.1175 0.1175 0.0137945 0.01175
Posterior nsnps H0 H1 H2 H3 H4
1.175000e+03 3.819485e-38 4.037077e-01 3.703500e-39 3.858714e-02 5.577051e-01 Results fail decision rule H4 > 0.9
Results fail decision rule H4 > 0.9Getting data for rs10899970.
Subset required data.
Check temp1 data.Warning in coloc::check_dataset(temp1, warn.minp = 1e-10) :
minimum p value is: 5.4509e-09
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.
Check temp2 data.Warning in coloc::check_dataset(temp2, warn.minp = 1e-10) :
minimum p value is: 0.00014931
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.
Plot required data.
Warning in check_dataset(d = dataset2, 2) :
minimum p value is: 0.00014931
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.PP.H0.abf PP.H1.abf PP.H2.abf PP.H3.abf PP.H4.abf
0.001320 0.848000 0.000167 0.107000 0.043600
[1] "PP abf for shared variant: 4.36%"
Colocalization.Coloc analysis of trait 1, trait 2
SNP Priors p1 p2 p12
1e-04 1e-04 1e-05
Hypothesis Priors H0 H1 H2 H3 H4
0.6284776 0.1641 0.1641 0.0269124 0.01641
Posterior nsnps H0 H1 H2 H3 H4
1.641000e+03 1.319568e-03 8.476892e-01 1.669259e-04 1.071894e-01 4.363490e-02 Results fail decision rule H4 > 0.9
Results fail decision rule H4 > 0.9Getting data for rs9633535.
Subset required data.
Check temp1 data.Warning in coloc::check_dataset(temp1, warn.minp = 1e-10) :
minimum p value is: 3.7901e-07
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.
Check temp2 data.Warning in coloc::check_dataset(temp2, warn.minp = 1e-10) :
minimum p value is: 1.7844e-05
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.
Plot required data.
Warning in check_dataset(d = dataset2, 2) :
minimum p value is: 1.7844e-05
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.PP.H0.abf PP.H1.abf PP.H2.abf PP.H3.abf PP.H4.abf
0.1540 0.5240 0.0155 0.0527 0.2540
[1] "PP abf for shared variant: 25.4%"
Colocalization.Coloc analysis of trait 1, trait 2
SNP Priors p1 p2 p12
1e-04 1e-04 1e-05
Hypothesis Priors H0 H1 H2 H3 H4
0.8188193 0.083 0.083 0.0068807 0.0083
Posterior nsnps H0 H1 H2 H3 H4
830.00000000 0.15368622 0.52399743 0.01553863 0.05272534 0.25405239 Results fail decision rule H4 > 0.9
Results fail decision rule H4 > 0.9Getting data for rs10762577.
Subset required data.
Check temp1 data.Warning in coloc::check_dataset(temp1, warn.minp = 1e-10) :
minimum p value is: 1.2432e-06
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.
Check temp2 data.Warning in coloc::check_dataset(temp2, warn.minp = 1e-10) :
minimum p value is: 0.00073363
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.
Plot required data.
Warning in check_dataset(d = dataset1, 1) :
minimum p value is: 1.2432e-06
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.
Warning in check_dataset(d = dataset2, 2) :
minimum p value is: 0.00073363
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.PP.H0.abf PP.H1.abf PP.H2.abf PP.H3.abf PP.H4.abf
0.06080 0.85200 0.00167 0.02340 0.06210
[1] "PP abf for shared variant: 6.21%"
Colocalization.Coloc analysis of trait 1, trait 2
SNP Priors p1 p2 p12
1e-04 1e-04 1e-05
Hypothesis Priors H0 H1 H2 H3 H4
0.9440809 0.0263 0.0263 0.00068906 0.00263
Posterior nsnps H0 H1 H2 H3 H4
2.630000e+02 6.077064e-02 8.521307e-01 1.669798e-03 2.335196e-02 6.207694e-02 Results fail decision rule H4 > 0.9
Results fail decision rule H4 > 0.9Getting data for rs11063120.
Subset required data.
Check temp1 data.Warning in coloc::check_dataset(temp1, warn.minp = 1e-10) :
minimum p value is: 9.1265e-09
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.
Check temp2 data.Warning in coloc::check_dataset(temp2, warn.minp = 1e-10) :
minimum p value is: 3.7022e-05
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.
Plot required data.
Warning in check_dataset(d = dataset2, 2) :
minimum p value is: 3.7022e-05
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.PP.H0.abf PP.H1.abf PP.H2.abf PP.H3.abf PP.H4.abf
0.00446 0.70700 0.00105 0.16600 0.12200
[1] "PP abf for shared variant: 12.2%"
Colocalization.Coloc analysis of trait 1, trait 2
SNP Priors p1 p2 p12
1e-04 1e-04 1e-05
Hypothesis Priors H0 H1 H2 H3 H4
0.7992586 0.0916 0.0916 0.0083814 0.00916
Posterior nsnps H0 H1 H2 H3 H4
9.160000e+02 4.464724e-03 7.069361e-01 1.047960e-03 1.658103e-01 1.217409e-01 Results fail decision rule H4 > 0.9
Results fail decision rule H4 > 0.9Getting data for rs9515203.
Subset required data.
Check temp1 data.Warning in coloc::check_dataset(temp1, warn.minp = 1e-10) :
minimum p value is: 3.1936e-07
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.
Check temp2 data.Warning in coloc::check_dataset(temp2, warn.minp = 1e-10) :
minimum p value is: 0.0005347
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.
Plot required data.
Warning in check_dataset(d = dataset2, 2) :
minimum p value is: 0.0005347
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.PP.H0.abf PP.H1.abf PP.H2.abf PP.H3.abf PP.H4.abf
0.1080 0.7390 0.0142 0.0972 0.0417
[1] "PP abf for shared variant: 4.17%"
Colocalization.Coloc analysis of trait 1, trait 2
SNP Priors p1 p2 p12
1e-04 1e-04 1e-05
Hypothesis Priors H0 H1 H2 H3 H4
0.686176 0.1401 0.1401 0.019614 0.01401
Posterior nsnps H0 H1 H2 H3 H4
1.401000e+03 1.078718e-01 7.390042e-01 1.419994e-02 9.723877e-02 4.168536e-02 Results fail decision rule H4 > 0.9
Results fail decision rule H4 > 0.9Getting data for rs7182103.
Subset required data.
Check temp1 data.Warning in coloc::check_dataset(temp1, warn.minp = 1e-10) :
minimum p value is: 1.9223e-09
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.
Check temp2 data.Warning in coloc::check_dataset(temp2, warn.minp = 1e-10) :
minimum p value is: 5.9197e-06
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.
Plot required data.
Warning in check_dataset(d = dataset2, 2) :
minimum p value is: 5.9197e-06
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.PP.H0.abf PP.H1.abf PP.H2.abf PP.H3.abf PP.H4.abf
9.93e-05 8.11e-01 1.82e-05 1.49e-01 3.97e-02
[1] "PP abf for shared variant: 3.97%"
Colocalization.Coloc analysis of trait 1, trait 2
SNP Priors p1 p2 p12
1e-04 1e-04 1e-05
Hypothesis Priors H0 H1 H2 H3 H4
0.7836873 0.0984 0.0984 0.00967272 0.00984
Posterior nsnps H0 H1 H2 H3 H4
9.840000e+02 9.928622e-05 8.113958e-01 1.820615e-05 1.487462e-01 3.974052e-02 Results fail decision rule H4 > 0.9
Results fail decision rule H4 > 0.9Getting data for rs7412.
Subset required data.
Check temp1 data.
Check temp2 data.Warning in coloc::check_dataset(temp2, warn.minp = 1e-10) :
minimum p value is: 2.9412e-06
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.
Plot required data.
Warning in check_dataset(d = dataset2, 2) :
minimum p value is: 2.9412e-06
If this is what you expected, this is not a problem.
If this is not as small as you expected, please check the 02_data vignette.PP.H0.abf PP.H1.abf PP.H2.abf PP.H3.abf PP.H4.abf
0.01090 0.64900 0.00402 0.23800 0.09770
[1] "PP abf for shared variant: 9.77%"
Colocalization.Coloc analysis of trait 1, trait 2
SNP Priors p1 p2 p12
1e-04 1e-04 1e-05
Hypothesis Priors H0 H1 H2 H3 H4
0.7896519 0.0958 0.0958 0.00916806 0.00958
Posterior nsnps H0 H1 H2 H3 H4
9.580000e+02 1.094517e-02 6.491241e-01 4.018346e-03 2.382180e-01 9.769439e-02 Results fail decision rule H4 > 0.9
Results fail decision rule H4 > 0.9
Summary colocalization
ENPP1/ENPP3 locus
File: res_EA_vs_AA_single.1.6_131595002_132595002.txt
c(nsnps = 891, PP.H0.abf = 0.212726180436945, PP.H1.abf = 0.0690743673677815, PP.H2.abf = 0.0214723764600455, PP.H3.abf = 0.00628185511784578, PP.H4.abf = 0.690445220617383)
Conclusion: 69.0% probability that the locus is shared between both ancestries and includes the same causal variant.
IGFBP3 locus
File: res_EA_vs_AA_single.2.7_45460645_46460645.txt
c(nsnps = 1147, PP.H0.abf = 0.598769555553722, PP.H1.abf = 0.108030020950978, PP.H2.abf = 0.0685581712636257, PP.H3.abf = 0.0121567818199419, PP.H4.abf = 0.212485470411732)
Conclusion: 59.9% probability that the locus is not associated in either ancestry, but 10.8 that it is European-specific, and 21.2% that it is shared between both ancestries and includes the same causal variant.
CXCL12 locus
File: res_EA_vs_AA_single.3.10_44015716_45334720.txt
c(nsnps = 1641, PP.H0.abf = 0.00131956785769608, PP.H1.abf = 0.847689205708678, PP.H2.abf = 0.000166925872986017, PP.H3.abf = 0.107189395862512, PP.H4.abf = 0.0436349046981281)
Conclusion: 84.8% probability that the locus is European-specific, but 4.3-10.7% probability that the locus is shared between both ancestries.
ARID5B locus
File: res_EA_vs_AA_single.4.10_63336088_64336088.txt
c(nsnps = 830, PP.H0.abf = 0.153686217856268, PP.H1.abf = 0.52399742579072, PP.H2.abf = 0.0155386299541433, PP.H3.abf = 0.0527253377600635, PP.H4.abf = 0.254052388638805)
Conclusion: 52.4% probability that the locus is European-specific, but 25.4% probability that the locus is shared between both ancestries and includes different causal variants.
ADK locus
File: res_EA_vs_AA_single.5.10_75417431_76417431.txt
c(nsnps = 263, PP.H0.abf = 0.0607706367611815, PP.H1.abf = 0.85213066483438, PP.H2.abf = 0.00166979761323014, PP.H3.abf = 0.0233519569734565, PP.H4.abf = 0.0620769438177522)
Conclusion: 85.2% probability that the locus is European-specific, but 2.3-6.2% probability that the locus is shared between both ancestries.
FGF23 locus
File: res_EA_vs_AA_single.6.12_3986618_4986618.txt
c(nsnps = 916, PP.H0.abf = 0.00446472426354749, PP.H1.abf = 0.706936091086602, PP.H2.abf = 0.00104796032447378, PP.H3.abf = 0.165810337258493, PP.H4.abf = 0.121740887066883)
Conclusion: 70.7% probability that the locus is European-specific, but 12.2-16.6% probability that the locus is shared between both ancestries.
COL4A1/COL4A2 locus
File: res_EA_vs_AA_single.7.13_110549623_111549623.txt
c(nsnps = 1401, PP.H0.abf = 0.10787176608552, PP.H1.abf = 0.739004164794562, PP.H2.abf = 0.0141999398787617, PP.H3.abf = 0.0972387717170162, PP.H4.abf = 0.0416853575241402)
Conclusion: 73.9% probability that the locus is European-specific, but 4.1-9.7% probability that the locus is shared between both ancestries.
MORF4L locus
File: res_EA_vs_AA_single.8.15_78623946_79623946.txt
c(nsnps = 984, PP.H0.abf = 9.92862156724414e-05, PP.H1.abf = 0.811395804353311, PP.H2.abf = 1.82061468585998e-05, PP.H3.abf = 0.148746181818894, PP.H4.abf = 0.0397405214652622)
Conclusion: 81.1% probability that the locus is European-specific, but 14.9% probability that the locus is shared between both ancestries and includes different causal variants.
PHACTR1 locus
File: res_EA_vs_AA_single.9.6_12403957_13403957.txt
c(nsnps = 1082, PP.H0.abf = 2.64349916808974e-21, PP.H1.abf = 0.440110764390371, PP.H2.abf = 1.85706249346447e-22, PP.H3.abf = 0.0303883523691947, PP.H4.abf = 0.529500883240437)
Conclusion: 44.0% probability that the locus is European-specific, but 52.9% probability that the locus is shared between both ancestries and includes the same causal variant.
CDKN2A/CDKN2B locus
File: res_EA_vs_AA_single.10.9_21624744_22624744.txt
c(nsnps = 1175, PP.H0.abf = 3.8194848005248e-38, PP.H1.abf = 0.40370772152903, PP.H2.abf = 3.70349960775562e-39, PP.H3.abf = 0.0385871394254695, PP.H4.abf = 0.557705139045494)
Conclusion: 40.4% probability that the locus is European-specific, but 55.8% probability that the locus is shared between both ancestries and includes the same causal variant.
APOE locus
File: res_EA_vs_AA_single.11.19_44912079_45912079.txt
c(nsnps = 958, PP.H0.abf = 0.0109451663563229, PP.H1.abf = 0.649124092491782, PP.H2.abf = 0.00401834569685491, PP.H3.abf = 0.238218007619045, PP.H4.abf = 0.0976943878359947)
Conclusion: 64.9% probability that the locus is European-specific, but 23.8% probability that the locus is shared between both ancestries and includes different causal variants.
Overarching conclusions:
- The European-specific analysis is better powered than the African-American-specific analysis
- Taking into account the issues of power, it is more likely that the majority of loci are shared between ancestries. The ADK and COL4A1/COL4A2 loci are the least shared between ancestries, and both analyses are well-_under_powered for the IGFBP3 locus.
- Given sufficient and more equal sample sizes between ancestries more firm conclusions could be drawn, but applying Occam’s razor it is more likely that these loci are shared between ancestries than not.
Session information
Version: v1.4.2
Last update: 2022-06-28
Written by: Sander W. van der Laan (s.w.vanderlaan-2[at]umcutrecht.nl).
Description: Script to create plot regional association plots.
Minimum requirements: R version 3.4.3 (2017-06-30) -- 'Single Candle', Mac OS X El Capitan
Changes log
* v1.4.2 Added formal quantification of colocalization between ancestries.
* v1.4.1 Added mirror plots and scatter plots.
* v1.4.0 Update with AA data.
* v1.3.0 Added the credible sets to the aditional regions.
* v1.2.0 Added in aditional regions.
* v1.1.0 Created PNG and PDF of top loci regions.
* v1.0.0 Initial version. sessionInfo()Saving environment
save.image(paste0(PROJECT_loc, "/",Today,".",PROJECTNAME,".RegionalAssociationPlots.RData"))| © 1979-2022 Sander W. van der Laan | s.w.vanderlaan[at]gmail.com | swvanderlaan.github.io. |
---
title: "Regional association plotting of 11 loci associated with CAC."
author: "[Sander W. van der Laan, PhD](https://swvanderlaan.github.io) | @swvanderlaan | s.w.vanderlaan@gmail.com"
date: "`r Sys.Date()`"
output:
  html_notebook:
    cache: yes
    code_folding: hide
    collapse: yes
    df_print: paged
    fig.align: center
    fig_caption: yes
    fig_height: 6
    fig_retina: 2
    fig_width: 7
    highlight: tango
    theme: lumen
    toc: yes
    toc_float:
      collapsed: no
      smooth_scroll: yes
mainfont: Arial
subtitle: "A 'druggable-MI-targets' project"
editor_options:
  chunk_output_type: inline
---

# Setup
We will clean the environment, setup the locations, define colors, and create a datestamp.

_Clean the environment._
```{r echo = FALSE}
# rm(list = ls())

```

_Set locations and working directories..._
```{r LocalSystem, echo = FALSE}
source("scripts/local.system.R")

```

_... a package-installation function ..._
```{r}
source("scripts/functions.R")
```


_... and load those packages._
```{r loading_packages, message=FALSE, warning=FALSE}
source("scripts/packages05.R")

```

_We will create a datestamp and define the Utrecht Science Park Colour Scheme_.
```{r Setting: Colors}

Today = format(as.Date(as.POSIXlt(Sys.time())), "%Y%m%d")
Today.Report = format(as.Date(as.POSIXlt(Sys.time())), "%A, %B %d, %Y")

source("scripts/colors.R")

```

```{r global_options, include = FALSE}
# further define some knitr-options.
knitr::opts_chunk$set(fig.width = 12, fig.height = 8, fig.path = 'Figures/', 
                      wwarning = TRUE, # show warnings during codebook generation
  message = TRUE, # show messages during codebook generation
  error = TRUE, # do not interrupt codebook generation in case of errors,
                # usually better for debugging
  echo = TRUE,  # show R code
                      eval = TRUE)
ggplot2::theme_set(ggplot2::theme_minimal())
pander::panderOptions("table.split.table", Inf)
```


# Introduction

We will parse the data to create regional association plots for each of the 11 loci. 

# Setting the NPG colors

```{r}
library("scales")
pal_npg("nrc")(10)
show_col(pal_npg("nrc")(10))

# show_col(pal_npg("nrc", alpha = 0.6)(10))

```


# Load data

## European and African-American

We need to load the meta-analysis summary statistics from the European - African-American ancestry analysis first.
```{r}

gwas_sumstats_racer_EA_AA <- readRDS(file = paste0(OUT_loc, "/gwas_sumstats_complete_racer.EA_AA.rds"))

```

## European-only

We need to load the meta-analysis summary statistics from the European-only ancestry analysis.
```{r}

# gwas_sumstats_EA <- fread(paste0(GWAS_loc,"/CAC1000G_EA_FINAL_FUMA.txt.gz"),
#                          showProgress = TRUE)
# names(gwas_sumstats_EA)[names(gwas_sumstats_EA) == "Pos"] <- "Position"
# saveRDS(gwas_sumstats_EA, file = paste0(OUT_loc, "/gwas_sumstats.EA.rds"))
# 
# gwas_sumstats_racer_EA <- subset(gwas_sumstats_EA,
#                               select = c("MarkerName", "rsID", "Chr", "Position", "Pvalue"))
# 
# saveRDS(gwas_sumstats_racer_EA, file = paste0(OUT_loc, "/gwas_sumstats_racer.EA.rds"))
# rm(gwas_sumstats_EA)

gwas_sumstats_racer_EA <- readRDS(file = paste0(OUT_loc, "/gwas_sumstats_racer.EA.rds"))

```


## African-American-only

We need to load the meta-analysis summary statistics from the African-American-only ancestry analysis.
```{r}
# gwas_sumstats_unfiltered_AA <- fread(paste0(GWAS_loc,"/CAC1000G_AA_FINAL_FUMA.unfiltered.txt.gz"),
#                          showProgress = TRUE)
# saveRDS(gwas_sumstats_unfiltered_AA, file = paste0(OUT_loc, "/gwas_sumstats_unfiltered.AA.rds"))

# gwas_sumstats_racer_unfiltered_AA <- subset(gwas_sumstats_unfiltered_AA,
#                               select = c("MarkerName", "rsID", "Chr", "Position", "Pvalue"))
# 
# saveRDS(gwas_sumstats_racer_unfiltered_AA, file = paste0(OUT_loc, "/gwas_sumstats_unfiltered_racer.AA.rds"))
# 
# gwas_sumstats_racer_unfiltered_AA <- readRDS(file = paste0(OUT_loc, "/gwas_sumstats_unfiltered_racer.AA.rds"))

# 
# gwas_sumstats_AA <- fread(paste0(GWAS_loc,"/CAC1000G_AA_FINAL_FUMA.filtered.txt.gz"),
#                          showProgress = TRUE)
# saveRDS(gwas_sumstats_AA, file = paste0(OUT_loc, "/gwas_sumstats.AA.rds"))
# 
# gwas_sumstats_racer_AA <- subset(gwas_sumstats_AA,
#                               select = c("MarkerName", "rsID", "Chr", "Position", "Pvalue"))
# 
# saveRDS(gwas_sumstats_racer_AA, file = paste0(OUT_loc, "/gwas_sumstats_racer.AA.rds"))
# rm(gwas_sumstats_AA)
gwas_sumstats_racer_AA <- readRDS(file = paste0(OUT_loc, "/gwas_sumstats_racer.AA.rds"))

```

# Regional association plotting: EU-AA-ancestry

## Top 11 loci

We are interested in 11 top loci. We will plot these using the EU-AA-ancestry data.

```{r}
library(openxlsx)
variant_list <- read.xlsx(paste0(TARGET_loc, "/Variants.xlsx"), sheet = "TopLoci")

DT::datatable(variant_list)

```


### All loci
Let's do some plotting.

```{r}
library(RACER)
# Make directory for plots
ifelse(!dir.exists(file.path(PROJECT_loc, "/RACER")), 
       dir.create(file.path(PROJECT_loc, "/RACER")), 
       FALSE)
RACER_loc = paste0(PROJECT_loc,"/RACER")

variants_of_interest <- c(variant_list$rsID)

variants_of_interest_fewgenes <- c("rs9349379")#, "rs3844006", "rs2854746", "rs4977575", "rs9633535", "rs11063120", "rs9515203", "rs7182103")

for(VARIANT in variants_of_interest){
  cat(paste0("Getting data for ", VARIANT,".\n"))

  tempCHR <- subset(variant_list, rsID == VARIANT)[,5]
  tempSTART <- subset(variant_list, rsID == VARIANT)[,17]
  tempEND <- subset(variant_list, rsID == VARIANT)[,18]
  tempVARIANTnr <- subset(variant_list, rsID == VARIANT)[,1]

  cat("\nSubset required data.\n")
  temp <- subset(gwas_sumstats_racer_EA_AA, Chr == tempCHR & (Position >= tempSTART & Position <= tempEND))
  
  cat("\nFormatting association data.\n")
  temp_f = RACER::formatRACER(assoc_data = temp, chr_col = 3, pos_col = 4, p_col = 5)

  cat("\nGetting LD data.\n")
  temp_f_ld = RACER::ldRACER(assoc_data = temp_f, rs_col = 2, pops = "EUR", lead_snp = VARIANT)
  
  cat(paste0("\nPlotting region surrounding ", VARIANT," on ",tempCHR,":",tempSTART,"-",tempEND,".\n"))
  p1 <- singlePlotRACER2(assoc_data = temp_f_ld, 
                               chr = tempCHR, build = "hg19", 
                               plotby = "snp", snp_plot = VARIANT,
                               label_lead = TRUE, gene_track_h = 2, gene_name_s = 1.75)
  
  print(p1)
  cat(paste0("Saving image for ", VARIANT,".\n"))
  # ggsave(filename = paste0(RACER_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.png"), plot = last_plot())
  # ggsave(filename = paste0(RACER_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.pdf"), plot = last_plot())
  # ggsave(filename = paste0(RACER_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.eps"), plot = last_plot())

  # rm(temp, p1,
  #    temp_f, temp_f_ld,
  #    tempCHR, tempSTART, tempEND,
  #    VARIANT, tempVARIANTnr)
  
}


```

### Loci with many genes
These are genetic loci with many genes. 

```{r}

variants_of_interest_manygenes <- c("rs7412", "rs10762577")

for(VARIANT in variants_of_interest_manygenes){
  cat(paste0("Getting data for ", VARIANT,".\n"))

  tempCHR <- subset(variant_list, rsID == VARIANT)[,5]
  tempSTART <- subset(variant_list, rsID == VARIANT)[,17]
  tempEND <- subset(variant_list, rsID == VARIANT)[,18]
  tempVARIANTnr <- subset(variant_list, rsID == VARIANT)[,1]

  cat("\nSubset required data.\n")
  temp <- subset(gwas_sumstats_racer_EA_AA, Chr == tempCHR & (Position >= tempSTART & Position <= tempEND))
  
  cat("\nFormatting association data.\n")
  temp_f = RACER::formatRACER(assoc_data = temp, chr_col = 3, pos_col = 4, p_col = 5)

  cat("\nGetting LD data.\n")
  temp_f_ld = RACER::ldRACER(assoc_data = temp_f, rs_col = 2, pops = "EUR", lead_snp = VARIANT)
  
  cat(paste0("\nPlotting region surrounding ", VARIANT," on ",tempCHR,":",tempSTART,"-",tempEND,".\n"))
  p1 <- singlePlotRACER2(assoc_data = temp_f_ld, 
                               chr = tempCHR, build = "hg19", 
                               plotby = "snp", snp_plot = VARIANT,
                               label_lead = TRUE, gene_track_h = 0.75, gene_name_s = 1.75)
  
  print(p1)
  cat(paste0("Saving image for ", VARIANT,".\n"))
  # ggsave(filename = paste0(RACER_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.png"), plot = last_plot())
  # ggsave(filename = paste0(RACER_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.pdf"), plot = last_plot())
  # ggsave(filename = paste0(RACER_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.eps"), plot = last_plot())
  
  rm(temp, p1,
     temp_f, temp_f_ld,
     tempCHR, tempSTART, tempEND,
     VARIANT, tempVARIANTnr)
  
}
```

### CXCL12

The _CXCL12_ genetic locus.

```{r}
variants_of_interest_cxcl12 <- c("rs10899970")

for(VARIANT in variants_of_interest_cxcl12){
  cat(paste0("Getting data for ", VARIANT,".\n"))

  tempCHR <- subset(variant_list, rsID == VARIANT)[,5]
  tempSTART <- subset(variant_list, rsID == VARIANT)[,17]
  tempEND <- subset(variant_list, rsID == VARIANT)[,18]
  tempVARIANTnr <- subset(variant_list, rsID == VARIANT)[,1]

  cat("\nSubset required data.\n")
  temp <- subset(gwas_sumstats_racer_EA_AA, Chr == tempCHR & (Position >= tempSTART & Position <= tempEND))
  
  cat("\nFormatting association data.\n")
  temp_f = RACER::formatRACER(assoc_data = temp, chr_col = 3, pos_col = 4, p_col = 5)

  cat("\nGetting LD data.\n")
  temp_f_ld = RACER::ldRACER(assoc_data = temp_f, rs_col = 2, pops = "EUR", lead_snp = VARIANT)
  
  cat(paste0("\nPlotting region surrounding ", VARIANT," on ",tempCHR,":",tempSTART,"-",tempEND,".\n"))
  p1 <- singlePlotRACER2(assoc_data = temp_f_ld, 
                               chr = tempCHR, build = "hg19", set = "all",
                               plotby = "snp", snp_plot = VARIANT,
                               label_lead = TRUE, gene_track_h = 0.75, gene_name_s = 1.75)
  
  print(p1)
  cat(paste0("Saving image for ", VARIANT,".\n"))
  # ggsave(filename = paste0(RACER_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.png"), plot = last_plot())
  # ggsave(filename = paste0(RACER_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.pdf"), plot = last_plot())
  # ggsave(filename = paste0(RACER_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.eps"), plot = last_plot())
  
  rm(temp, p1,
     temp_f, temp_f_ld,
     tempCHR, tempSTART, tempEND,
     VARIANT, tempVARIANTnr)
  
}
```

## Additional regional plots

### Listing regions of interest

We want to create some regional association plots to combine with teh UCSC browser tracks, thus we need the exact same regions. 

```{r}
library(openxlsx)
add_list <- read.xlsx(paste0(TARGET_loc, "/Variants.xlsx"), sheet = "AdditionalPlots")

DT::datatable(add_list)

```

### Credible Sets

We want to color the credible sets, which we load here.

```{r}
credset <- as_tibble(fread(paste0(PROJECT_loc, "/CredibleSets/CAC_EUR_AFR_cred_set_all_loci_50kb.txt")))

credset
```

### Combining GWAS with Credible Set

We want to add the posterior probabilities and make a variable to color by.
```{r}

gwas_sumstats_racer_credset <- merge(gwas_sumstats_racer_EA_AA, 
                                     credset %>% select(RSID, Posterior_Prob), 
                                     sort = FALSE,
                                     by.x = "rsID", by.y = "RSID", all.x = TRUE) %>%
  # mutate(., Posterior_Prob = ifelse(is.na(Posterior_Prob), 0, Posterior_Prob)) %>%
  mutate(CredSet = case_when(Posterior_Prob > 0 ~ '95% credible set',
                             TRUE ~ 'not in credible set'))

head(gwas_sumstats_racer_credset)

table(gwas_sumstats_racer_credset$CredSet)

summary(gwas_sumstats_racer_credset$Posterior_Prob)

```

### Plotting

```{r}
library(RACER)
# library(plotly)

# Make directory for plots
ifelse(!dir.exists(file.path(PROJECT_loc, "/RACER")), 
       dir.create(file.path(PROJECT_loc, "/RACER")), 
       FALSE)
RACER_loc = paste0(PROJECT_loc,"/RACER")

variants_of_interest <- c(add_list$rsID)


for(VARIANT in variants_of_interest){
  cat(paste0("Getting data for ", VARIANT,".\n"))

  tempCHR <- subset(add_list, rsID == VARIANT)[,4]
  tempSTART <- subset(add_list, rsID == VARIANT)[,5]
  tempEND <- subset(add_list, rsID == VARIANT)[,6]
  tempNAME <- subset(add_list, rsID == VARIANT)[,3]

  cat("\nSubset required data.\n")
  temp <- subset(gwas_sumstats_racer_credset, Chr == tempCHR & (Position >= tempSTART & Position <= tempEND))
  
  cat("\nFormatting association data.\n")
  temp_f = RACER::formatRACER(assoc_data = temp, chr_col = 3, pos_col = 4, p_col = 5)

  cat("\nGetting LD data.\n")
  # temp_f_ld = RACER::ldRACER(assoc_data = temp_f, rs_col = 2, pops = "EUR", lead_snp = VARIANT)
  
  cat(paste0("\nPlotting region surrounding ", VARIANT," on ",tempCHR,":",tempSTART,"-",tempEND,".\n"))
  
  p1 <- singlePlotRACER2(assoc_data = temp_f, 
                         chr = tempCHR, build = "hg19", 
                         plotby = "coord", snp_plot = VARIANT,
                         start_plot = tempSTART, end_plot = tempEND,
                         label_lead = FALSE, 
                         grey_colors = FALSE, 
                         cred_set = TRUE, 
                         gene_track_h = 3, gene_name_s = 1.75)
  
  print(p1)
  
  cat(paste0("Saving image for ", VARIANT,".\n"))
  # ggsave(filename = paste0(RACER_loc, "/", tempNAME, ".", Today, ".",VARIANT,".",tempSTART,".",tempEND,".regional_assoc.png"), plot = p1)
  # ggsave(filename = paste0(RACER_loc, "/", tempNAME, ".", Today, ".",VARIANT,".",tempSTART,".",tempEND,".regional_assoc.pdf"), plot = p1)
  # ggsave(filename = paste0(RACER_loc, "/", tempNAME, ".", Today, ".",VARIANT,".",tempSTART,".",tempEND,".regional_assoc.eps"), plot = p1)

  # print(ggplotly(p1))
  rm(temp, p1,
     temp_f,
     tempCHR, tempSTART, tempEND,
     VARIANT, tempNAME)
}
```


## Regional association plots in African-American

Note here that we plot the region, and not based on the lead variant of the EA-AA analyses.

```{r}
library(RACER)
# Make directory for plots
ifelse(!dir.exists(file.path(PROJECT_loc, "/RACER_AA")), 
       dir.create(file.path(PROJECT_loc, "/RACER_AA")), 
       FALSE)
RACER_AA_loc = paste0(PROJECT_loc,"/RACER_AA")

variants_of_interest <- c(variant_list$rsID)

variants_of_interest_fewgenes <- c("rs9349379", 
                                   "rs3844006", # throws an error which I don't understand immediately - could be that the variant is not present in AA 1000G data
                                   "rs2854746", "rs4977575", 
                                   "rs10899970",
                                   "rs9633535", # this one throws an LD error
                                   "rs10762577",
                                   "rs11063120", "rs9515203", "rs7182103",
                                   "rs7412")

for(VARIANT in variants_of_interest){
  cat(paste0("Getting data for ", VARIANT,".\n"))

  tempCHR <- subset(variant_list, rsID == VARIANT)[,5]
  tempSTART <- subset(variant_list, rsID == VARIANT)[,17]
  tempEND <- subset(variant_list, rsID == VARIANT)[,18]
  tempVARIANTnr <- subset(variant_list, rsID == VARIANT)[,1]

  cat("\nSubset required data.\n")
  temp <- subset(gwas_sumstats_racer_AA, Chr == tempCHR & (Position >= tempSTART & Position <= tempEND))

  cat("\nFormatting association data.\n")
  temp_f = RACER::formatRACER(assoc_data = temp, chr_col = 3, pos_col = 4, p_col = 5)
  
  # cat("\nGetting LD data.\n")
  temp_f_ld = RACER::ldRACER(assoc_data = temp_f, rs_col = 2, pops = "AFR", lead_snp = VARIANT)

  cat(paste0("\nPlotting region surrounding ", VARIANT," on ",tempCHR,":",tempSTART,"-",tempEND,".\n"))

  p1 <- singlePlotRACER2(assoc_data = temp_f, 
                         chr = tempCHR, build = "hg19", 
                         plotby = "coord", snp_plot = VARIANT,
                         start_plot = tempSTART, end_plot = tempEND,
                         label_lead = FALSE, gene_track_h = 2, gene_name_s = 1.75)
  
  print(p1)
  cat(paste0("Saving image for ", VARIANT,".\n"))
  ggsave(filename = paste0(RACER_AA_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.AA.png"), plot = last_plot())
  ggsave(filename = paste0(RACER_AA_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.AA.pdf"), plot = last_plot())
  ggsave(filename = paste0(RACER_AA_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_assoc.AA.eps"), plot = last_plot())

  rm(temp, p1,
     temp_f, temp_f_ld,
     tempCHR, tempSTART, tempEND,
     VARIANT, tempVARIANTnr)
  
}

```

# Approximate Bayes Factor colocalisation analyses

The idea behind the *Approximate Bayes Factor (ABF)* analysis is that the association of each trait with SNPs in a region may be summarised by a vector of 0s and at most a single 1, with the 1 indicating the causal SNP (so, assuming a single causal SNP for each trait). 

The posterior probability of each possible configuration can be calculated and so, crucially, can the posterior probabilities that the traits share their configurations. This allows us to estimate the support for the following cases, i.e. hypotheses:

𝐻0: neither trait has a genetic association in the region
𝐻1: only trait 1 has a genetic association in the region
𝐻2: only trait 2 has a genetic association in the region
𝐻3: both traits are associated, but with different causal variants
𝐻4: both traits are associated and share a single causal variant

To what extent do the loci between European and African-American ancestries overlap? We are working on the assumption that the 11 loci are _the_ loci and will test whether these overlap, _i.e._ colocalize.

## Preparation

Let's make sure we have `remotes` and `coloc` installed. 
```{r}
if(!require("remotes"))
   install.packages("remotes") # if necessary
library(remotes)
install_github("chr1swallace/coloc@main",
               build_vignettes = TRUE)

library(coloc)

```

## Visualization

We can create mirror and scatter plot for each region. 

```{r}

library(RACER)
# Make directory for plots
ifelse(!dir.exists(file.path(PROJECT_loc, "/RACER_EA_vs_AA")), 
       dir.create(file.path(PROJECT_loc, "/RACER_EA_vs_AA")), 
       FALSE)
RACER_EA_vs_AA_loc = paste0(PROJECT_loc,"/RACER_EA_vs_AA")

variants_of_interest <- c(variant_list$rsID)

# variants_of_interest_fewgenes <- c("rs9349379", 
#                                    "rs3844006", # throws an error which I don't understand immediately - could be that the variant is not present in AA 1000G data
#                                    "rs2854746", "rs4977575", 
#                                    "rs10899970",
#                                    "rs9633535", 
#                                    "rs10762577",
#                                    "rs11063120", "rs9515203", "rs7182103",
#                                    "rs7412")
variants_of_interest_fewgenes <- c("rs10762577",
                                   "rs11063120", "rs9515203", "rs7182103",
                                   "rs7412")
for(VARIANT in variants_of_interest){
  cat(paste0("Getting data for ", VARIANT,".\n"))

  tempCHR <- subset(variant_list, rsID == VARIANT)[,5]
  tempSTART <- subset(variant_list, rsID == VARIANT)[,17]
  tempEND <- subset(variant_list, rsID == VARIANT)[,18]
  tempVARIANTnr <- subset(variant_list, rsID == VARIANT)[,1]

  cat("\nSubset required data.\n")
  temp1 <- subset(gwas_sumstats_racer_EA, Chr == tempCHR & (Position >= tempSTART & Position <= tempEND))
  temp2 <- subset(gwas_sumstats_racer_AA, Chr == tempCHR & (Position >= tempSTART & Position <= tempEND))

  cat("\nFormatting association data.\n")
  temp_f1 = RACER::formatRACER(assoc_data = temp1, chr_col = 3, pos_col = 4, p_col = 5)
  temp_f2 = RACER::formatRACER(assoc_data = temp2, chr_col = 3, pos_col = 4, p_col = 5)
  
  # cat("\nGetting LD data.\n")
  temp_f_ld1 = RACER::ldRACER(assoc_data = temp_f1, rs_col = 2, pops = "EUR", lead_snp = VARIANT)
  temp_f_ld2 = RACER::ldRACER(assoc_data = temp_f2, rs_col = 2, pops = "AFR", lead_snp = VARIANT)

  cat(paste0("\nPlotting region surrounding ", VARIANT," on ",tempCHR,":",tempSTART,"-",tempEND,".\n"))

  p1 <- mirrorPlotRACER(assoc_data1 = temp_f_ld1, 
                        assoc_data2 = temp_f_ld2, 
                        chr = tempCHR, 
                        name1 = "European ancestry", 
                         name2 = "African-American ancestry", 
                        plotby = "coord", 
                        start_plot = tempSTART, end_plot = tempEND)
  
  print(p1)
  
  p2 <- scatterPlotRACER(assoc_data1 = temp_f_ld1, 
                         assoc_data2 = temp_f_ld2, 
                         chr = tempCHR, 
                         name1 = "European ancestry", 
                         name2 = "African-American ancestry", 
                         region_start = tempSTART, 
                         region_end = tempEND, 
                         ld_df = 1)
  
  print(p2)
  
  cat(paste0("Saving image for ", VARIANT,".\n"))
  ggsave(filename = paste0(RACER_EA_vs_AA_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_mirror.EA_vs_AA.png"), plot = p1)
  ggsave(filename = paste0(RACER_EA_vs_AA_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_mirror.EA_vs_AA.pdf"), plot = p1)
  ggsave(filename = paste0(RACER_EA_vs_AA_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_mirror.EA_vs_AA.eps"), plot = p1)

  cat(paste0("Saving image for ", VARIANT,".\n"))
  ggsave(filename = paste0(RACER_EA_vs_AA_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_scatter.EA_vs_AA.png"), plot = p2)
  ggsave(filename = paste0(RACER_EA_vs_AA_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_scatter.EA_vs_AA.pdf"), plot = p2)
  ggsave(filename = paste0(RACER_EA_vs_AA_loc, "/", tempVARIANTnr, ".", Today, ".",VARIANT,".regional_scatter.EA_vs_AA.eps"), plot = p2)

  
  rm(temp1, temp2,
     p1,p2,
     temp_f1, temp_f_ld1,
     temp_f2, temp_f_ld2,
     tempCHR, tempSTART, tempEND,
     VARIANT, tempVARIANTnr)
  
}


```

## Quantify colocalization

We want to quantify the overlap. 

Setting up an output directory for `coloc`.

```{r}
ifelse(!dir.exists(file.path(PROJECT_loc, "/COLOC_EA_vs_AA")), 
       dir.create(file.path(PROJECT_loc, "/COLOC_EA_vs_AA")), 
       FALSE)
COLOC_EA_vs_AA_loc = paste0(PROJECT_loc,"/COLOC_EA_vs_AA")

```


### Data colleaction

Preparing data for `coloc`: we require beta and standard errors for `coloc`.

First the European ancestry data, next the African-American ancestry data.

```{r}
gwas_sumstats_EA <- fread(paste0(GWAS_loc,"/CAC1000G_EA_FINAL_FUMA.txt.gz"),
                         showProgress = TRUE)
names(gwas_sumstats_EA)[names(gwas_sumstats_EA) == "Pos"] <- "Position"

gwas_sumstats_racer_EA <- subset(gwas_sumstats_EA,
                              select = c("MarkerName", "rsID", "Chr", "Position", 
                                         "Effect", "StdErr",
                                         "Pvalue",
                                         "N"))
rm(gwas_sumstats_EA)

```

```{r}
gwas_sumstats_AA <- fread(paste0(GWAS_loc,"/CAC1000G_AA_FINAL_FUMA.filtered.txt.gz"),
                         showProgress = TRUE)
names(gwas_sumstats_AA)[names(gwas_sumstats_AA) == "Pos"] <- "Position"
names(gwas_sumstats_AA)[names(gwas_sumstats_AA) == "SE"] <- "StdErr"

gwas_sumstats_racer_AA <- subset(gwas_sumstats_AA,
                              select = c("MarkerName", "rsID", "Chr", "Position", 
                                         "Effect", "StdErr",
                                         "Pvalue",
                                         "N"))
rm(gwas_sumstats_AA)

```

### Colocalization

Now we are reading to formally test the colocalization per trait. Note that `trait 1` = European ancestry; `trait 2` = African-American ancestry.

```{r}
for(VARIANT in variants_of_interest){
  cat(paste0("Getting data for ", VARIANT,".\n"))

  tempCHR <- subset(variant_list, rsID == VARIANT)[,5]
  tempSTART <- subset(variant_list, rsID == VARIANT)[,17]
  tempEND <- subset(variant_list, rsID == VARIANT)[,18]
  tempVARIANTnr <- subset(variant_list, rsID == VARIANT)[,1]

  cat("\nSubset required data.\n")
  temp1 <- subset(gwas_sumstats_racer_EA, Chr == tempCHR & (Position >= tempSTART & Position <= tempEND))
  temp2 <- subset(gwas_sumstats_racer_AA, Chr == tempCHR & (Position >= tempSTART & Position <= tempEND))

  cat("\nCheck temp1 data.\n")
  temp1 <- rename_with(temp1, tolower)
  
  # correcting column names
  temp1 <- rename(temp1, beta = effect)
  temp1 <- rename(temp1, se = stderr)
  temp1 <- rename(temp1, snp = rsid)
  
  # calculating things
  temp1$varbeta <- temp1$se^2
  
  temp1 <- as.list(temp1) # critical, as coloc expects a list of variables
  
  temp1$type <- "quant"
  temp1$sdY <- 1

  coloc::check_dataset(temp1, warn.minp = 1e-10)
  
  cat("\nCheck temp2 data.\n")
  temp2 <- rename_with(temp2, tolower)
  
  # correcting column names
  temp2 <- rename(temp2, beta = effect)
  temp2 <- rename(temp2, se = stderr)
  temp2 <- rename(temp2, snp = rsid)
  
  # calculating things
  temp2$varbeta <- temp2$se^2
  
  temp2 <- as.list(temp2) # critical, as coloc expects a list of variables
  
  temp2$type <- "quant"
  temp2$sdY <- 1

  coloc::check_dataset(temp2, warn.minp = 1e-10)
  
  cat("\nPlot required data.\n")
  plot_dataset(temp1)
  plot_dataset(temp2)

  res_temp1_vs_temp2_single <- coloc.abf(dataset1 = temp1, 
                                         dataset2 = temp2)

  cat("\nColocalization.\n")
  print(res_temp1_vs_temp2_single)
  write_lines(res_temp1_vs_temp2_single, file = paste0(COLOC_EA_vs_AA_loc, "/res_EA_vs_AA_single.",
                                                       tempVARIANTnr,".",tempCHR,"_",tempSTART,"_",tempEND,".txt"))

  coloc::sensitivity(res_temp1_vs_temp2_single, "H4 > 0.9")
  
  # Step 1: Call the pdf command to start the plot
  pdf(file = paste0(COLOC_EA_vs_AA_loc, "/res_EA_vs_AA_single.",
                    tempVARIANTnr,".",tempCHR,"_",tempSTART,"_",tempEND,".pdf"))   # The directory you want to save the file in
  # Step 2: Create the plot with R code
  coloc::sensitivity(res_temp1_vs_temp2_single, "H4 > 0.9")

  # Step 3: Run dev.off() to create the file!
  dev.off()
  
  rm(temp1, temp2,
     # p1,p2,
     # temp_f1, temp_f_ld1,
     # temp_f2, temp_f_ld2,
     tempCHR, tempSTART, tempEND,
     VARIANT, tempVARIANTnr)
  
}

```

### Summary colocalization

_ENPP1_/_ENPP3_ locus

File: res_EA_vs_AA_single.1.6_131595002_132595002.txt

c(nsnps = 891, PP.H0.abf = 0.212726180436945, PP.H1.abf = 0.0690743673677815, PP.H2.abf = 0.0214723764600455, PP.H3.abf = 0.00628185511784578, PP.H4.abf = 0.690445220617383)

**Conclusion: 69.0% probability that the locus is shared between both ancestries and includes the same causal variant.**


_IGFBP3_ locus

File: res_EA_vs_AA_single.2.7_45460645_46460645.txt

c(nsnps = 1147, PP.H0.abf = 0.598769555553722, PP.H1.abf = 0.108030020950978, PP.H2.abf = 0.0685581712636257, PP.H3.abf = 0.0121567818199419, PP.H4.abf = 0.212485470411732)

**Conclusion: 59.9% probability that the locus is not associated in either ancestry, but 10.8 that it is European-specific, and 21.2% that it is shared between both ancestries and includes the same causal variant.**


_CXCL12_ locus

File: res_EA_vs_AA_single.3.10_44015716_45334720.txt

c(nsnps = 1641, PP.H0.abf = 0.00131956785769608, PP.H1.abf = 0.847689205708678, PP.H2.abf = 0.000166925872986017, PP.H3.abf = 0.107189395862512, PP.H4.abf = 0.0436349046981281)

**Conclusion: 84.8% probability that the locus is European-specific, but 4.3-10.7% probability that the locus is shared between both ancestries.**


_ARID5B_ locus

File: res_EA_vs_AA_single.4.10_63336088_64336088.txt

c(nsnps = 830, PP.H0.abf = 0.153686217856268, PP.H1.abf = 0.52399742579072, PP.H2.abf = 0.0155386299541433, PP.H3.abf = 0.0527253377600635, PP.H4.abf = 0.254052388638805)

**Conclusion: 52.4% probability that the locus is European-specific, but 25.4% probability that the locus is shared between both ancestries and includes different causal variants.**


_ADK_ locus

File: res_EA_vs_AA_single.5.10_75417431_76417431.txt

c(nsnps = 263, PP.H0.abf = 0.0607706367611815, PP.H1.abf = 0.85213066483438, PP.H2.abf = 0.00166979761323014, PP.H3.abf = 0.0233519569734565, PP.H4.abf = 0.0620769438177522)

**Conclusion: 85.2% probability that the locus is European-specific, but 2.3-6.2% probability that the locus is shared between both ancestries.**


_FGF23_ locus

File: res_EA_vs_AA_single.6.12_3986618_4986618.txt

c(nsnps = 916, PP.H0.abf = 0.00446472426354749, PP.H1.abf = 0.706936091086602, PP.H2.abf = 0.00104796032447378, PP.H3.abf = 0.165810337258493, PP.H4.abf = 0.121740887066883)

**Conclusion: 70.7% probability that the locus is European-specific, but 12.2-16.6% probability that the locus is shared between both ancestries.**


_COL4A1_/_COL4A2_ locus

File: res_EA_vs_AA_single.7.13_110549623_111549623.txt

c(nsnps = 1401, PP.H0.abf = 0.10787176608552, PP.H1.abf = 0.739004164794562, PP.H2.abf = 0.0141999398787617, PP.H3.abf = 0.0972387717170162, PP.H4.abf = 0.0416853575241402)

**Conclusion: 73.9% probability that the locus is European-specific, but 4.1-9.7% probability that the locus is shared between both ancestries.**


_MORF4L_ locus

File: res_EA_vs_AA_single.8.15_78623946_79623946.txt

c(nsnps = 984, PP.H0.abf = 9.92862156724414e-05, PP.H1.abf = 0.811395804353311, PP.H2.abf = 1.82061468585998e-05, PP.H3.abf = 0.148746181818894, PP.H4.abf = 0.0397405214652622)

**Conclusion: 81.1% probability that the locus is European-specific, but 14.9% probability that the locus is shared between both ancestries and includes different causal variants.**


_PHACTR1_ locus

File: res_EA_vs_AA_single.9.6_12403957_13403957.txt

c(nsnps = 1082, PP.H0.abf = 2.64349916808974e-21, PP.H1.abf = 0.440110764390371, PP.H2.abf = 1.85706249346447e-22, PP.H3.abf = 0.0303883523691947, PP.H4.abf = 0.529500883240437)

**Conclusion: 44.0% probability that the locus is European-specific, but 52.9% probability that the locus is shared between both ancestries and includes the same causal variant.**


_CDKN2A_/_CDKN2B_ locus

File: res_EA_vs_AA_single.10.9_21624744_22624744.txt

c(nsnps = 1175, PP.H0.abf = 3.8194848005248e-38, PP.H1.abf = 0.40370772152903, PP.H2.abf = 3.70349960775562e-39, PP.H3.abf = 0.0385871394254695, PP.H4.abf = 0.557705139045494)

**Conclusion: 40.4% probability that the locus is European-specific, but 55.8% probability that the locus is shared between both ancestries and includes the same causal variant.**


_APOE_ locus

File: res_EA_vs_AA_single.11.19_44912079_45912079.txt

c(nsnps = 958, PP.H0.abf = 0.0109451663563229, PP.H1.abf = 0.649124092491782, PP.H2.abf = 0.00401834569685491, PP.H3.abf = 0.238218007619045, PP.H4.abf = 0.0976943878359947)

**Conclusion: 64.9% probability that the locus is European-specific, but 23.8% probability that the locus is shared between both ancestries and includes different causal variants.**


Overarching conclusions:

1. The European-specific analysis is better powered than the African-American-specific analysis
2. Taking into account the issues of power, it is more likely that the majority of loci are shared between ancestries. The _ADK_ and _COL4A1_/_COL4A2_ loci are the least shared between ancestries, and both analyses are well-_under_powered for the _IGFBP3_ locus. 
3. Given sufficient and more equal sample sizes between ancestries more firm conclusions could be drawn, but applying Occam's razor it is more likely that these loci are shared between ancestries than not.

# Session information

------

    Version:      v1.4.2
    Last update:  2022-06-28
    Written by:   Sander W. van der Laan (s.w.vanderlaan-2[at]umcutrecht.nl).
    Description:  Script to create plot regional association plots.
    Minimum requirements: R version 3.4.3 (2017-06-30) -- 'Single Candle', Mac OS X El Capitan
    
    Changes log
    * v1.4.2 Added formal quantification of colocalization between ancestries.
    * v1.4.1 Added mirror plots and scatter plots.
    * v1.4.0 Update with AA data.
    * v1.3.0 Added the credible sets to the aditional regions.
    * v1.2.0 Added in aditional regions.
    * v1.1.0 Created PNG and PDF of top loci regions.
    * v1.0.0 Initial version. 

------

```{r eval = TRUE}
sessionInfo()

```


# Saving environment
```{r Saving}
save.image(paste0(PROJECT_loc, "/",Today,".",PROJECTNAME,".RegionalAssociationPlots.RData"))

```


------
<sup>&copy; 1979-2022 Sander W. van der Laan | s.w.vanderlaan[at]gmail.com | [swvanderlaan.github.io](https://swvanderlaan.github.io).</sup>
------

  
